_________________
ARTICLE
_________________
DYNAMIC INCORPORATION OF FOREIGN LAW
MICHAEL C. DORF
†
Lawmaking bodies in one polity sometimes incorporate the law of another
polity “dynamically,” so that when the law of the foreign jurisdiction changes,
the law of the incorporating jurisdiction changes automatically. Dynamic incorporation can save lawmaking costs, lead to better legal rules and standards,
and solve collective action problems. Thus, the phenomenon is widespread.
Dynamic incorporation does, however, delegate lawmaking power. Further, as
the formal and practical barriers to revocation of the act of dynamic incorporation become higher, that act comes closer to a cession of sovereignty, and for democratic polities, such cessions entail a democratic loss. Accordingly, dynamic
incorporation of foreign law has proven controversial both within federal systems and at the international level. The problem is most acute when nation
states agree to delegate lawmaking power to a supranational entity. In order to
gain the reciprocal benefits of cooperation and coordination, the delegation
must be functionally irrevocable or nearly so. Representation of the member nation states within the decision-making structures of the supranational entity
can ameliorate, but cannot fully compensate for, the resulting democratic losses
†
Robert S. Stevens Professor of Law, Cornell University Law School. I am very
grateful for the helpful suggestions, comments, and questions of Ittai Bar-Siman-Tov,
Perry Bechky, George Bermann, Ashutosh Bhagwat, Richard Briffault, Richard Brooks,
Neil Buchanan, Christina Burnett, Sherry Colb, Perry Dane, Michael Doyle, Ariela
Dubler, Merritt Fox, Katherine Franke, Jane Ginsburg, Suzanne Goldberg, Victor
Goldberg, Jeffrey Gordon, Kent Greenawalt, Vicki Jackson, Mitchell Kane, Avery Katz,
Thomas Merrill, Gillian Metzger, Henry Monaghan, Trevor Morrison, Alex Raskolnikov, Joseph Raz, Charles Sabel, Wojciech Sadurski, Carol Sanger, David Schizer,
Christopher Schroeder, Jonathan Siegel, Allan Stein, Peter Strauss, William Stuntz,
Edward Swaine, and John Witt, and for the analytic and research assistance of Eric
Chesin, Jordan Connors, and Jessica Karp.
(103)
University of Pennsylvania Law Review
104
[Vol. 157: 103
suffered by those nation states. More broadly, the benefits of dynamic incorporation must always be balanced against its costs, including the cost to selfgovernance.
INTRODUCTION ......................................................................................104
I.
HOW DYNAMIC INCORPORATION THREATENS DEMOCRACY ...........115
A. The Prima Facie Threat to Democracy.....................................115
B. The Revocability Spectrum .....................................................118
II. DIRECTIONS OF INCORPORATION: UP, DOWN, AND ACROSS .........124
A. Up........................................................................................124
B. Down ...................................................................................125
C. Across...................................................................................127
III. REASONS FOR DYNAMIC INCORPORATION ......................................132
A. Avoiding Lawmaking Costs...................................................133
B. Accommodating Local Diversity Through
Customization.......................................................................138
C. Coordination and Reciprocity ................................................146
IV. THE POLITICAL SAFEGUARDS OF DYNAMIC INCORPORATION .........152
A. Representation as Remedy ......................................................153
B. Representative Courts? ..........................................................158
CONCLUSION..........................................................................................168
INTRODUCTION
When lawmaking bodies incorporate by reference laws from other
jurisdictions, typically they do not thereby delegate any lawmaking
power. Incorporation by reference functions as a shorthand. It
adopts the law as it stands at the moment of incorporation. Future
changes in the law of the adopted jurisdiction do not take effect in the
adopting jurisdiction, unless and until the lawmaking body in the
adopting jurisdiction takes the further step of incorporating the
changes. Thus, incorporation by reference is usually static. However,
lawmaking bodies sometimes employ a strategy of dynamic incorpora1
tion of foreign law, so that if and when the law of the incorporated
1
Throughout this Article, I use the term “foreign” to refer to truly foreign sovereigns, transnational bodies, and different sovereigns within the American federal system. In doing so, I do not mean to deny that for many purposes “[i]nternational law is
part of our law.” The Paquete Habana, 175 U.S. 677, 700 (1900). But I am interested
in how and when international law becomes part of domestic law, and so it would be
question-begging to begin with this point. Likewise, I do not deny that the relation of
U.S. states to one another and to the national government differs in key respects from
the relation of a U.S. state or the national government to a truly foreign sovereign. See
2008]
Dynamic Incorporation
105
jurisdiction changes, the law of the incorporating jurisdiction changes
with it. In contrast to static incorporation, dynamic incorporation
2
does delegate lawmaking authority, and has therefore proven controversial.
At the national level, scholars disagree about whether the Constitution permits the United States to enter international agreements
(whether by treaty or by other means) that cede to foreign, partly foreign, or transnational bodies the power to make rules of law that are
self-executing within the United States and thus binding on U.S. government officials. For example, David Golove and Henry Monaghan
each argue that since the founding, the federal government has had
the power to enter agreements authorizing international and foreign
bodies to take legislative, executive, and adjudicatory actions that not
only bind the United States as a matter of international obligation, but
3
also operate internally. In addition, Edward Swaine has offered a
functional justification of most delegations of lawmaking authority to
4
transnational bodies. By contrast, Curtis Bradley, John Yoo, and others offer a different picture of the historical record and contend that,
in any event, modern understandings of the Constitution limit the
ability of the federal government to place the making, execution, and
interpretation of law in the hands of foreign bodies that it does not
5
control.
Testa v. Katt, 330 U.S. 386, 389-90 (1947) (explaining that, in light of the Supremacy
Clause of Article VI, a state cannot treat federal law as “foreign” in the way that it treats
the law of a truly foreign sovereign). But cf. Michigan v. Long, 463 U.S. 1032, 1068
(1983) (Stevens, J., dissenting) (likening Michigan to Finland, with respect to the federal government).
2
See Jonathan R. Siegel, The Use of Legislative History in a System of Separated Powers,
53 VAND. L. REV. 1457, 1480 (2000) (“Incorporation by reference of pre-existing text
cannot violate the nondelegation doctrine, because it does not give away any power.”).
3
David Golove, The New Confederalism: Treaty Delegations of Legislative, Executive, and
Judicial Authority, 55 STAN. L. REV. 1697 (2003); Henry Paul Monaghan, Article III and
Supranational Judicial Review, 107 COLUM. L. REV. 833 (2007). Whether any given international obligation in fact operates internally of its own force depends on whether
the obligation-creating instrument “conveys an intention that it be ‘self-executing’ and
[whether the treaty] is ratified on these terms.” Medellin v. Texas, 128 S. Ct. 1346,
1356 (2008) (internal quotation marks omitted) (quoting Igartúa-De La Rosa v.
United States, 417 F.3d 145, 150 (1st Cir. 2005) (en banc)).
4
See Edward T. Swaine, The Constitutionality of International Delegations, 104 COLUM.
L. REV. 1492, 1613-14 (2004) (explaining that diffusing authority among multiple
transnational bodies often serves the same values as federalism).
5
See, e.g., Curtis A. Bradley, International Delegations, the Structural Constitution, and
Non-Self-Execution, 55 STAN. L. REV. 1557, 1596 (2003) (endorsing a non-self-execution
approach in order to “reduce many of the constitutional concerns associated with international delegations without significantly affecting the United States’s ability to par-
106
University of Pennsylvania Law Review
[Vol. 157: 103
The proper resolution of the federal constitutional question has
important consequences both for international agreements into which
the United States has already entered, such as the North American
6
Free Trade Agreement (NAFTA), and for international agreements
into which the United States may enter in the future, such as the
Kyoto Protocol and successor environmental treaties. If and when the
validity of such federal commitments reaches the United States Supreme Court, the Justices will no doubt be attentive to the constitutional text, structure, and original understanding, as well as historical
practice, their own precedents, and the expected consequences of
7
whatever rule they announce.
Likewise, similar tensions will need to be resolved in accordance
with the particular language, history, and interpretive conventions in
the European Union (EU) and its member states. The European
Court of Justice insists that treaty signatories must bring their domestic law—including constitutional provisions—into conformity with EU
ticipate in international institutions”); Michael J. Glennon & Allison R. Hayward, Collective Security and the Constitution: Can the Commander in Chief Power Be Delegated to the
United Nations?, 82 GEO. L.J. 1573, 1598-99 (1994) (calling the delegation of Congress’s
war power to the United Nations Security Council constitutionally “troublesome”);
Julian G. Ku, The Delegation of Federal Power to International Organizations: New Problems
with Old Solutions, 85 MINN. L. REV. 71, 76-77 (2000) (arguing that treaty-based international delegations are unconstitutional due to accountability and legitimacy deficiencies); John C. Yoo, Kosovo, War Powers, and the Multilateral Future, 148 U. PA. L. REV.
1673, 1708-20 (2000) (contending that placing U.S. troops under foreign command
“creates significant tension with American constitutional principles of government accountability and popular sovereignty, as promoted by the Appointments Clause, the
unitary executive, and the non-delegation doctrine”); John C. Yoo, The New Sovereignty
and the Old Constitution: The Chemical Weapons Convention and the Appointments Clause, 15
CONST. COMMENT. 87, 88-89, 128 (1998) (arguing that the Chemical Weapons Convention is in tension with the Appointments Clause insofar as it “permits individuals
who are not officers of the national government to exercise authority under federal
law that affects the rights of American citizens”).
6
U.S.-Can.-Mex., Dec. 8-17, 1992, 32 I.L.M. 289 (1993). Under Chapter 19 of
NAFTA, each signatory nation retains the right to apply its domestic antidumping and
countervailing-duty law, subject to review by “binational panels.” Id. ch. 19. It is unclear whether the decisions of these binational panels, which are not subject to review
by an Article III court, see 19 U.S.C. § 1516a(g) (2006), can directly bind federal agencies. See Michael C. Dorf, Fallback Law, 107 COLUM. L. REV. 303, 313-14 (2007) (noting
“doubts about the constitutionality” of a binational panel system).
7
Although there is no official list, the factors identified in the text closely resemble the “modalities” of constitutional interpretation identified by Philip Bobbitt. See
PHILIP BOBBITT, CONSTITUTIONAL INTERPRETATION 12-13 (1991) (listing history, text,
structure, consequences, precedent, and national ethos).
Dynamic Incorporation
2008]
107
8
obligations, while member states insist that EU law must yield to con9
trary domestic constitutional requirements where they conflict.
To be sure, the looming collision in the EU concerns static as well
as dynamic incorporation: under the Westphalian approach of some
national constitutional courts, even an EU norm that was clear at the
time of accession would have to yield to a contrary national constitu-
8
See, e.g., Case 11/70, Internationale Handelsgesellschaft mbH v. Einfuhr-und
Vorratsstelle für Getreide und Futtermittel, 1970 E.C.R. 1125 (“[T]he validity of a
Community measure . . . cannot be affected by allegations that it runs counter to . . .
the constitution of that State . . . .”); Case 6/64, Costa v. ENEL, 1964 E.C.R. 585, 593
(“[The European Economic Community] Treaty . . . could not . . . be overridden by
domestic legal provisions . . . .”). The European Community “has formed only one of
three ‘pillars’ constituting the overarching European Union entity.” PAUL CRAIG &
GRÁINNE DE BÚRCA, EU LAW: TEXT, CASES AND MATERIALS 3 (3d ed. 2003). For an
argument that the well-established supremacy of European Community norms applies
for all European Union norms (including those of the second and third pillars), see
Koen Lenaerts & Tim Corthaut, Of Birds and Hedges: The Role of Primacy in Invoking
Norms of EU Law, 31 EUR. L. REV. 287, 289-90 (2006).
9
See, e.g., Wyrok z dnia 11 maja 2005 r. [ Judgment of May 11, 2005], K 18/04
(Pol.) (asserting the primacy of Polish constitutional law over EU law). A summary of
the “principal reasons for the ruling” can be found at the official website of the Polish
Constitutional Tribunal, see Trybunal Konstytucyjny [Pol. Constitutional Trib.], Poland’s Membership in the European Union (Accession Treaty), May 11, 2005,
http://www.trybunal.gov.pl/eng/summaries/documents/K_18_04_GB.pdf (last visited
Oct. 1, 2008), but only the Polish language text is authoritative. See also Krystyna
Kowalik-Banczyk, Should We Polish It Up? The Polish Constitutional Tribunal and the Idea of
Supremacy of EU Law, 6 GERMAN L.J. 1355, 1361-65 (2005), available at
http://www.germanlawjournal.com/pdf/Vol06No10/PDF_Vol_06_No_10_1355-1366_
Developments_Kowalik%20final.pdf (analyzing the Accession Treaty case).
National constitutional courts have relied on three lines of argument to reject the
supremacy of European law: the primacy of fundamental rights, the notion that some
European actions are ultra vires, and the supposed supremacy of national constitutions. See Mattias Kumm & Victor Ferreres Comella, The Primacy Clause of the Constitutional Treaty and the Future of Constitutional Conflict in the European Union, 3 INT’L J.
CONST. L. 473, 474-76 (2005) (identifying these lines of argument as three grounds on
which national constitutional courts have asserted authority to overrule EU law). The
fundamental-rights argument is expressed in the German Constitutional Court’s Solange (“so long as”) cases, which first rejected supremacy of European law on the
ground that it did not include fundamental rights protection. See Bundesverfassungsgericht [BVerfG] [Federal Constitutional Court] May 29, 1974, 37 Entscheidungen des
Bundesverfassungsgerichts [BVerfGE] 271 (F.R.G.) [hereinafter Solange I]. It then
accepted supremacy conditionally, finding that adequate rights protection had been
established. See BVerfG Oct. 22, 1986, 73 BVerfGE 339 (F.R.G.) [hereinafter Solange
II]; see also Wojciech Sadurski, “Solange, Chapter 3”: Constitutional Courts in Central
Europe—Democracy—European Union 2-26 (Dep’t of Law, Eur. Univ. Inst., Working Paper
No. 2006/40, 2006), available at http://ssrn.com/abstract=963757 (describing the use
of Solange-type arguments by constitutional courts in postcommunist Central and Eastern European EU member states).
University of Pennsylvania Law Review
108
[Vol. 157: 103
10
tional norm. But the issue is more acute with respect to dynamic incorporation. For preexisting EU norms, the act of ratification provides legitimacy for any subsequent sublimation of a national norm to
a European one. By contrast, where the relevant EU norm is promulgated (by the European Commission, say) after some nation’s accession, all that vindicates the EU norm is the original, perhaps decadesold, act of accession. Given the well-mooted “democratic deficit” in
11
the EU, dynamic incorporation of EU law thus means replacing domestic norms with supranational ones of questionable legitimacy.
The legitimacy of dynamic incorporation of foreign law also poses
potential difficulties within federal systems such as the United States,
both at the state and national levels. Although some state constitutions expressly prohibit static as well as dynamic incorporation by ref12
erence, courts have been reluctant to approve dynamic incorporation even in states that do not generally prohibit incorporation by
reference. Depending on how one counts, either twelve or fifteen
state high courts forbid dynamic incorporation of federal law as an
13
impermissible delegation of lawmaking power, and more might forbid it if they were to face the question directly.
10
See Kumm & Comella, supra note 9, at 474-76 (describing jurisprudence on the
supremacy of national constitutional law over EU law).
11
The term “democratic deficit” can be attributed to David Marquand. See DAVID
MARQUAND, PARLIAMENT FOR EUROPE 64-66 (1979) (explaining the democratic deficit
resulting from the lack of governmental accountability). For an overview of the democratic-deficit debate, see CRAIG & DE BÚRCA, supra note 8, at 167-75.
12
See, e.g., LA. CONST. art. III, § 15(B) (“No system or code of laws shall be
adopted by general reference to it.”); N.D. CONST. art. IV, § 13, cl. 4 (“No bill may
be . . . incorporated in any other bill by reference to its title only . . . .”); OKLA. CONST.
art. V, § 57 (“[N]o law shall be revived, amended, or the provisions thereof extended
or conferred, by reference to its title only . . . .”).
13
Twelve states clearly forbid dynamic incorporation: Arizona, see State v. Williams, 583 P.2d 251, 254-55 (Ariz. 1978) (“[I]t is universally held that an incorporation
by state statute of rules, regulations, and statutes of federal bodies to be promulgated
subsequent to the enactment of the state statute constitutes an unlawful delegation of
legislative power.”); California, see People v. Kruger, 121 Cal. Rptr. 581, 583-84 (Cal.
App. Dep’t Super. Ct. 1975) (citing Brock v. Superior Court, 71 P.2d 209, 212-13 (Cal.
1937)); Florida, see Fla. Indus. Comm’n v. State ex. rel. Orange State Oil Co., 21 So. 2d
599, 603 (Fla. 1945); Hawaii, see State v. Tengan, 691 P.2d 365, 373 (Haw. 1984);
Maine, see State v. Webber, 133 A. 738, 740 (Me. 1926); Michigan, see Lievense v. Mich.
Unemployment Comp. Comm’n, 55 N.W.2d 857, 859 (Mich. 1952); Ohio, see State v.
Gill, 584 N.E.2d 1200, 1201-02 (Ohio 1992); Oregon, see Seale v. McKennon, 336 P.2d
340, 345 (Or. 1959); South Carolina, see Santee Mills v. Query, 115 S.E. 202, 205-06
(S.C. 1922); South Dakota, see State v. Johnson, 173 N.W.2d 894, 895 (S.D. 1970) (per
curiam); Washington, see State v. Dougall, 570 P.2d 135, 137-39 (Wash. 1977) (en
banc); and West Virginia, see State v. Grinstead, 206 S.E.2d 912, 920 (W. Va. 1974).
2008]
Dynamic Incorporation
109
Yet notwithstanding the skeptical states, domestic examples of dynamic incorporation of foreign law abound. For example, some states
define various terms in their income tax codes in a way that incorporates definitions found in the federal Internal Revenue Code, including definitions that Congress changes after the state decides to incor-
Arkansas, Kentucky, and Minnesota also forbid dynamic incorporation generally,
though apparently with exceptions. For Arkansas, compare Cheney v. St. Louis Sw. Ry.
Co., 394 S.W.2d 731, 733 (Ark. 1965) (“[A]ppellee’s tax liability to Arkansas is based
upon a formula subject to prospective federal legislation or administrative rules. It is
[therefore] unconstitutional.”), with Curry v. State, 649 S.W.2d 833, 836-37 (Ark. 1983)
(allowing dynamic incorporation of federal drug regulations where a state agency retains power to veto the incorporation); for Kentucky, compare Dawson v. Hamilton,
314 S.W.2d 532, 535 (Ky. 1958) (“[A]doption . . . of prospective Federal legislation . . .
constitutes an unconstitutional delegation of legislative power.” (internal quotation
marks omitted)), with Hamilton v. City of Louisville, 332 S.W.2d 539, 543 (Ky. 1960)
(noting that the dynamic incorporation of federal income tax provisions would probably not be unconstitutional); and for Minnesota, compare Wallace v. Comm’r of Taxation, 184 N.W.2d 588, 593 (Minn. 1971) (interpreting the incorporation of federal tax
law statically because the Minnesota legislature “could not . . . grant to Congress the
right to make future . . . changes in Minnesota law”), with State v. King, 257 N.W.2d
693, 697 (Minn. 1977) (upholding a state law that dynamically incorporated the federal determination of what constitutes a controlled substance).
Twelve state constitutions explicitly permit dynamic incorporation of federal tax
law. See infra note 14. The following seven state high courts have permitted, either expressly or impliedly but clearly, dynamic incorporation of federal law without relying
on an explicit constitutional provision: Alaska, see Hickel v. Stevenson, 416 P.2d 236,
239 & n.7 (Alaska 1966) (following Alaska S.S. Co. v. Mullaney, 180 F.2d 805, 816 (9th
Cir. 1950) in upholding the dynamic incorporation of federal tax law); Idaho, see State
v. Kellogg, 568 P.2d 514, 517-18 (Idaho 1977) (upholding a law criminalizing the unauthorized sale of drugs that require a prescription under state or federal law); Maryland, see Katzenberg v. Comptroller of the Treasury, 282 A.2d 465, 470, 473 (Md. 1971)
(upholding the dynamic incorporation of federal tax law); Massachusetts, see Parker
Affiliated Cos. v. Dep’t of Revenue, 415 N.E.2d 825, 831 (Mass. 1981); Nebraska, see
Anderson v. Tiemann, 155 N.W.2d 322, 327 (Neb. 1967); Pennsylvania, see Commonwealth v. Warner Bros. Theatres, Inc., 27 A.2d 62, 63-64 (Pa. 1942) (upholding the dynamic incorporation of the federal definition of net income); and Tennessee, see
McFaddin v. Jackson, 738 S.W.2d 176, 180-182 (Tenn. 1987) (upholding the dynamic
incorporation of federal tax law). In addition, the New Jersey Supreme Court has
commented favorably on dynamic incorporation of federal food-packaging regulations. See State v. Hotel Bar Foods, Inc., 112 A.2d 726, 732-33 (N.J. 1955) (discussing
dynamic incorporation at length before finding it was not at issue in the case).
Four other states have upheld dynamic incorporation where a state agency retains
power to veto the incorporation: Alabama, see McCurley v. State (Ex parte McCurley),
390 So. 2d 25, 27-29 (Ala. 1980) (upholding a law requiring the state board of health
to control a substance once it becomes controlled under federal law, absent objection
by the state agency); Arkansas, see Curry, 649 S.W.2d at 836-37; Minnesota, see King, 257
N.W.2d at 697; Missouri, see State v. Thompson, 627 S.W.2d 298, 302-03 (Mo. 1982) (en
banc).
University of Pennsylvania Law Review
110
[Vol. 157: 103
14
porate federal law. In addition, liability in tort in some states may
depend on compliance with federal law, even where the relevant federal standards go into effect after the state rule (whether legislatively
15
or judicially created) incorporating them. Further, some states interpret their state constitutions in “lockstep” with the U.S. Supreme
Court’s interpretation of parallel provisions of the Federal Constitution, with the consequence that a change in U.S. Supreme Court constitutional jurisprudence can change the meaning of the state consti16
tution.
14
See, e.g., DEL. CODE ANN. tit. 30, § 1105 (2007) (“The entire taxable income of a
resident of this State shall be the federal adjusted gross income as defined in the laws
of the United States as the same are or shall become effective for any taxable year with
the modifications . . . provided in this subchapter.”). Twelve state constitutions expressly allow dynamic incorporation of federal tax law: Colorado, see COLO. CONST.
art. X, § 19; Hawaii, see HAW. CONST. art. VII, § 2; Illinois, see ILL. CONST. art. IX, § 3;
Kansas, see KAN. CONST. art. XI, § 11; Missouri, see MO. CONST. art. X, § 4(d); New Mexico, see N.M. CONST. art. IV, § 18; New York, see N.Y. CONST. art. III, § 22; North Dakota,
see N.D. CONST. art. X, § 3; Oklahoma, see OKLA. CONST. art. X, § 12; Oregon, see OR.
CONST. art. IV, § 2; Utah, see UTAH CONST. art. XIII, § 4; and Virginia, see VA. CONST.
art. IV, § 11. In addition, courts in four states have upheld dynamic incorporation of
federal tax law even in the absence of explicit constitutional authorization: Alaska, see
Hickel, 416 P.2d at 238-39; Maryland, see Leatherwood v. State, 435 A.2d 477, 479-80
(Md. Ct. Spec. App. 1981); Massachusetts, see Parker Affiliated Cos., 415 N.E.2d at 831;
and Tennessee, see McFaddin, 738 S.W.2d at 181-82.
15
Although states differ over whether violation of a federal statute constitutes negligence by itself, raises a presumption of negligence, or counts as evidence thereof, neither the federal nature of a duty nor the fact that the federal statutory or regulatory
duty arose after the state’s general tort rules disqualifies the duty for incorporation
under state law as negligence per se. See Paul Sherman, Use of Federal Statutes in State
Negligence Per Se Actions, 13 WHITTIER L. REV. 831, 877-83 (1992) (explaining and cataloguing the various approaches taken by states toward negligence per se cases).
16
See, e.g., Mitchell v. State, 818 P.2d 1163, 1165 (Alaska Ct. App. 1991) (applying
the dynamic lockstep approach); Mefford v. White, 770 N.E.2d 1251, 1260 (Ill. App.
Ct. 2002) (“Illinois courts typically apply the ‘lockstep’ doctrine, which dictates that
provisions of the Illinois Constitution should be construed in the same manner as similar provisions of the United States Constitution.”); see also Robert A. Schapiro, Judicial
Deference and Interpretive Coordinacy in State and Federal Constitutional Law, 85 CORNELL L.
REV. 656, 692-93 (2000) (citing evidence of, and reasons for, the lockstep approach).
Interestingly, although the Florida Constitution requires that its search-and-seizure
provision “be construed in conformity with the 4th Amendment to the United States
Constitution, as interpreted by the United States Supreme Court,” FLA. CONST. art. I,
§ 12, the Florida Supreme Court has consistently held that legislative dynamic incorporation of federal statutory law amounts to an unconstitutional delegation. See, e.g.,
Adoue v. State, 408 So. 2d 567, 570 (Fla. 1981) (“Any attempt to incorporate a law as
part of this state’s body of laws prior to its creation by the appropriate federal authority
is an unconstitutional delegation of the legislative power.”); Hutchins v. Mayo, 197 So.
495, 498 (Fla. 1940) (“We cannot accept the view that the state commission may make
binding rules to be promulgated by the federal bureau in the future.”).
2008]
Dynamic Incorporation
111
Likewise, federal law sometimes dynamically incorporates state
law. For example, under the Conformity Act, federal courts applied
17
the procedural law of the states in which they sat respectively. Federal Rule of Evidence 501 (which was enacted by an ordinary Act of
Congress) makes state privilege law applicable in federal court where
18
state law provides the rule of decision. In addition, the Assimilative
Crimes Act incorporates the criminal law of the state in which federal
19
land is located. The federal judiciary provides yet another example:
in exercising its power to fashion federal common law in discrete areas of federal concern, the courts presumptively define the content of
20
federal law as state law.
The foregoing examples reveal that the federal government and
many states have found dynamic incorporation useful, even as other
states forbid the practice. But even without a blanket proscription on
dynamic incorporation, some instances of the phenomenon will appear problematic or at least highly peculiar. For example, to my
knowledge, no state has ever defined its law to dynamically incorporate the law of one of its sister states. Statutory efforts at harmonization, such as state-by-state adoption of restatements or uniform codes,
have typically functioned merely as a form of static incorporation, and
to the extent that state courts thereafter look to one another for guid21
ance, they treat out-of-state decisions only as persuasive precedent.
Moreover, to my knowledge, each example of federal dynamic incorporation of state law is a territorially limited accommodation to the
complexities created by a system of federalism. Congress has not attempted to make the law of a single state applicable, on a dynamic basis, to the nation as a whole; presumably an attempt to do so—for example, via a law specifying that federal law in admiralty cases shall be
17
See Conformity Act of 1872, ch. 255, § 6, 17 Stat. 196, 197, superseded by Rules
Enabling Act of 1934, 28 U.S.C. §§ 2071–77 (2006); see also 1 JAMES WM. MOORE ET AL.,
MOORE’S FEDERAL PRACTICE § 1.02 (3d ed. 2008) (“In 1872, Congress passed the socalled ‘Conformity Act,’ which required that federal district courts conform their procedure ‘as near as may be’ with that of the state in which the district was located.”).
18
FED. R. EVID. 501 (“[T]he privilege of a witness, person, government, State, or
political subdivision thereof shall be determined in accordance with State law.”).
19
18 U.S.C. § 13 (2006). The Act was upheld in United States v. Sharpnack, 355 U.S.
286, 294 (1958), in which the Court claimed that “Congress is within its constitutional
powers and legislative discretion when, after 123 years of experience with the policy of
conformity, it enacts that policy in its most complete and accurate form.”
20
See United States v. Kimbell Foods, Inc., 440 U.S. 715, 739-40 (1979) (expressing
a preference for the dynamic incorporation of state law rather than the formulation of
a federal rule, absent a demonstrated need for national uniformity).
21
See infra note 81 and accompanying text.
University of Pennsylvania Law Review
112
[Vol. 157: 103
Florida law or that federal law governing contracts with the federal
government shall be New York law—would prompt serious misgivings
if not objections on nondelegation or other constitutional grounds.
Whether the misgivings indeed rise to the level of state or federal
constitutional violations is not my main concern here. My goals are
analytic and general. This Article asks how and when dynamic incorporation of foreign law does or does not violate democratic principles.
Given the diversity of state and national constitutions, different states
and nations will reach different conclusions about whether and to
what extent dynamic incorporation of foreign law is permissible or desirable. Nor would it be surprising if some courts that were troubled
by some instances of dynamic incorporation nonetheless permitted
the practice on the grounds that the legislature was better situated to
22
make these judgments. To avoid paying undue attention to any particular constitutional text or set of doctrines, this Article treats dynamic incorporation of foreign law as a general question of institu23
tional design.
My analysis undoubtedly has implications for
constitutional interpretation, at least if one thinks that democratic
theory properly informs constitutional interpretation. But those implications are at most a bonus; my core concern is the institutionaldesign question as such.
This Article distinguishes between instances of dynamic incorporation that raise questions of sovereignty and those that, at most, raise
questions of delegation. The difference, which proves to be one of degree rather than kind, turns on revocability. When the several states
ratified the United States Constitution in 1789 and thereafter, they
made federal law—whatever its future content—irrevocably operative
24
in their respective territories. They thereby ceded some of their sovereignty. By contrast, should the United States by treaty authorize a
multinational body—the United Nations General Assembly, say—to
promulgate rules of law that are directly enforceable in U.S. courts as
22
Such an argument was made, for example, by the Kansas Supreme Court in upholding dynamic incorporation. See Mo. Pac. R.R. Co. v. McDonald, 486 P.2d 1347,
1352 (Kan. 1971) (“[The legislature] might have designated some other method but
the need and wisdom of its action is strictly a matter for legislative concern.”).
23
Reflecting the limits of my own expertise, I disproportionately draw examples
from the American context.
24
See Texas v. White, 74 U.S. (7 Wall.) 700, 725 (1869) (“The Constitution, in all
its provisions, looks to an indestructible Union, composed of indestructible States.”);
DANIEL FARBER, LINCOLN’S CONSTITUTION 77-81 (2003) (explaining Lincoln’s understanding of a perpetual, unbreakable Union).
2008]
Dynamic Incorporation
113
25
“the supreme Law of the Land,” no serious issue of sovereignty
would be raised. The United States would retain the power to abrogate the treaty either by its terms or, even failing that, by ordinary legislation pursuant to the rule that a later-in-time statute prevails over an
26
earlier-in-time treaty. Any objections would sound in principles of
nondelegation rather than sovereignty.
The distinction, however, between questions of sovereignty and
questions of delegation is not binary. These categories lie on a spectrum that includes intermediate cases, such as a partly entrenched
delegation. Indeed, even a completely unentrenched delegation is,
from a practical perspective, partly entrenched. For example, once a
treaty comes into effect, the burden of overcoming legislative inertia
to supersede it can be substantial because repealing a measure is al27
ways more difficult than not enacting it in the first place. Furthermore, as a practical matter, an act originally intended only as a delegation may become a cession of sovereignty over time, as arguably
occurred in the United States between the ratification of the Constitu28
tion and the conclusion of the Civil War.
To the extent that this Article advances a single thesis, it is this:
All acts of dynamic incorporation of foreign law pose a prima facie
threat to democratic principles, but as we move along the spectrum
from easily revocable delegations to irrevocable cessions of sovereignty, the burden of justification for dynamic incorporation increases. Further, representation of the power-delegating polity or its
members in the decision-making procedures of the body to which dynamic incorporation delegates power, can to some extent substitute
25
U.S. CONST. art. VI.
See, e.g., Whitney v. Robertson, 124 U.S. 190, 193-95 (1888) (holding that when a
treaty and statute conflict, “the one last in date will control the other”). To be sure, a
later-in-time U.S. law that conflicts with a UN Resolution might violate international
law, but that would not render the U.S. law inoperative internally.
27
For an excellent discussion of the degree to which “ordinary” legislation can
have entrenching effects, see Eric A. Posner & Adrian Vermeule, Essay, Legislative Entrenchment: A Reappraisal, 111 YALE L.J. 1665, 1686-88 (2002).
28
Justice Holmes put the point this way:
26
[W]hen we are dealing with words that also are a constituent act, like the Constitution of the United States, we must realize that they have called into life a
being the development of which could not have been foreseen completely by
the most gifted of its begetters. It was enough for them to realize or to hope
that they had created an organism; it has taken a century and has cost their
successors much sweat and blood to prove that they created a nation.
Missouri v. Holland, 252 U.S. 416, 433 (1920).
University of Pennsylvania Law Review
114
[Vol. 157: 103
for a fully satisfactory functional justification for delegation. Thus,
administrative convenience may be sufficient to justify a state’s dynamic incorporation of federal income tax law, where that incorporation can be undone by a simple legislative act. On the other hand,
something substantially stronger than administrative convenience
would be required for a polity to cede sovereignty.
The Article proceeds in four parts. Part I lays out the core of the
argument, explaining how dynamic incorporation threatens to undermine democratic principles and how, ceteris paribus, the threat
increases as the incorporating decision becomes progressively more
irrevocable.
Part II distinguishes among three sorts of dynamic incorporation:
upward incorporation, in which a political entity delegates lawmaking
authority to a larger political entity, as when a state dynamically incorporates federal law; downward incorporation, in which a political entity
delegates lawmaking authority to a subunit, as when the federal government dynamically incorporates state law; and horizontal incorporation, in which a political unit dynamically incorporates the law of a
comparable political unit, as when (to pick a hypothetical but plausible example), some state dynamically incorporates Delaware corpo29
rate law.
Part III catalogues and evaluates the principal sorts of reasons that
a polity might choose to dynamically incorporate the law of another
jurisdiction.
Part IV addresses the special case of irrevocable or nearly irrevocable upward incorporation. Upward incorporation will often prove
most attractive in circumstances in which states need assurances of reciprocity, but to make those assurances as meaningful as possible,
states will place limits on revocability. Thus, some of the most important instances of dynamic incorporation—states or nation states agreeing to be bound by decisions of interstate or supranational bodies—
prove the most problematic in terms of democratic principles, at least
29
I mostly bracket a fourth kind of delegation—to private actors—which raises
sufficiently distinct issues to warrant its own full treatment. Among these issues is the
question of how to distinguish, in a world without natural baselines, between delegations of government power to private actors and mere government failure to regulate
private power. For a useful discussion of the implications of privatization for the stateaction doctrine within the United States, see Gillian E. Metzger, Privatization as Delegation, 103 COLUM. L. REV. 1367 (2003). For an argument that beneficiaries of privatized
government programs have greater ability to enforce accountability than beneficiaries
of public programs, see Jody Freeman, The Private Role in Public Governance, 75 N.Y.U. L.
REV. 543, 646-66 (2000).
2008]
Dynamic Incorporation
115
prima facie. Part IV concludes that political representation of the
states or nation states in the interstate or supranational bodies may be
the best way to ameliorate the democracy-threatening character of
upward dynamic incorporation. It then asks whether such political
representation is a viable option when the foreign body whose lawmaking decisions are dynamically incorporated is a court.
The Article concludes by observing the kernel of truth in the argument of jurists and scholars who object to citation of foreign law by
U.S. courts interpreting U.S. law.
I. HOW DYNAMIC INCORPORATION THREATENS DEMOCRACY
This Part explains why and how, in a reasonably well-constituted
democracy, dynamic incorporation of foreign law can threaten democratic government. It then explains that the threat is roughly proportional to the obstacles to undoing the decision to dynamically incorporate foreign law. A truly irrevocable decision amounts to a
cession of sovereignty over the relevant subject matter, but even
nominally unentrenched decisions to dynamically incorporate foreign
law can be “sticky” in practice, and thus dynamic incorporation always
poses some prima facie threat to democratic values.
A. The Prima Facie Threat to Democracy
Dynamic incorporation of foreign law poses a prima facie threat
to the democracy of the incorporating polity because it takes decisions
out of the hands of the people’s representatives in that polity and
delegates them to persons and bodies that are accountable only to a
different polity, if at all. Under various circumstances, such a delegation of power may be sensible as a matter of policy. It may even increase the democratic accountability of the political system as a whole.
Nonetheless, where the polity that dynamically incorporates foreign
law is a reasonably well-constituted democracy, the act authorizing dynamic incorporation undermines self-government within that polity so
conceived.
In making this claim, I do not have in mind any special definitions
of terms like “democracy” or “reasonably well-constituted.” I mean
simply to distinguish between democratic and nondemocratic regimes
by whatever standards might be used to draw the distinction in clear
cases. For example, as of this writing, Canada and its provinces, the
United States and its individual states, and the member states of the
European Union all count as reasonably well-constituted democracies,
University of Pennsylvania Law Review
116
[Vol. 157: 103
30
while Cuba, Egypt, and North Korea do not. I am not concerned
31
with intermediate cases such as Russia. I want to limit my discussion
to reasonably well-constituted democracies because delegation of
power from the decision-making organs of nondemocracies, by definition, does not threaten democracy (although it may threaten other
values, such as national self-determination, at least if one assumes that
national self-determination does not depend on popular rule).
My contention that dynamic incorporation poses a prima facie
threat to democracy may seem inconsistent with the familiar notion
that within any political system as a whole—including, for these purposes, the international legal system—different kinds of decisions are
sensibly allocated to different levels of government. There must have
been an initial decision to design political institutions so that responsibility for various kinds of policy issues resides at the local, state, national, or supranational level, after all; unless we are prepared to say
that all such initial allocations of jurisdiction are perfect, we might
think that the decision of one polity to dynamically incorporate another’s law is simply a decision to revisit the initial allocation of authority among different levels of government. Is the resulting postdynamic-incorporation allocation necessarily inferior, from the standpoint of democracy, to the initial one?
Not necessarily, but from the perspective of the polity making the
decision, ceteris paribus, it is. To begin, where neither the incorporating polity nor its members are represented in the polity whose law
is dynamically incorporated, there is a clear democratic loss. To advert to categories that I develop more fully in Part II, dynamic incorporation is not even potentially equivalent to the reallocation of authority among different levels of government when it amounts to
horizontal incorporation or downward incorporation of a single jurisdiction’s law. Whatever else might be said for a decision by Newfoundland to be governed by Quebec law (horizontal incorporation), or for
Germany as a whole to be governed by the law of Bavaria on some sub30
See, e.g., LAZA KEKIC, ECONOMIST INTELLIGENCE UNIT, THE ECONOMIST INTELUNIT’S INDEX OF DEMOCRACY 3-5 tbl.1 (2007), available at
http://www.economist.com/media/pdf/DEMOCRACY_INDEX_2007_v3.pdf (last visited Oct. 1, 2008) (ranking countries from most to least “democratic”); Freedom
House, Freedom in the World 2008, http://www.freedomhouse.org/uploads/
fiw08launch/FIW08Tables.pdf (last visited Oct. 1, 2008) (ranking countries on “political rights” and “civil liberties”).
31
See KEKIC, supra note 30, at 4 tbl.1 (scoring Russia in the middle of the scale on
democracy); Freedom House, supra note 30 (placing Russia in the middle of the scale
on political rights and civil liberties).
LIGENCE
2008]
Dynamic Incorporation
117
ject (downward incorporation of a single jurisdiction’s law), in neither
case do we have a reallocation of power from one democratic unit to
another. So far as citizens of Newfoundland are concerned, the government of Quebec is not democratic because it does not represent
Newfoundlanders. Likewise, so far as Germany as a whole is concerned, the government of Bavaria is not democratic because it provides no representation whatsoever to persons residing in other
32
Länder (states).
By contrast, upward incorporation can, under some circumstances,
function as the rough equivalent of the reallocation of jurisdiction between government units, because the citizens of the incorporating jurisdiction have representation in the government of the jurisdiction
whose law is incorporated. Indeed, as Part IV argues, such representation can, under some circumstances, substitute for a fully effective
functional justification for dynamic incorporation.
Upward incorporation, however, invariably exacts a democratic
cost (albeit one that may sometimes be outweighed by countervailing
benefits). One need not believe that people always have the mobility
33
to find the polity with policies best suited to their preferences to recognize that, absent a geographically homogeneous distribution of policy preferences, people generally have a greater likelihood of approving the policy choices of their local governments than of their
supralocal governments. And whatever the distribution of preferences, as a matter of simple arithmetic, an individual’s vote is worth
proportionally more in a smaller polity than in a larger one (assuming
apportionment on principles that do not deviate wildly from one per34
son, one vote).
32
Nonetheless, as I acknowledge in Part II, horizontal incorporation and territorially limited downward incorporation can enhance democracy in the aggregate.
33
See Charles M. Tiebout, A Pure Theory of Local Expenditures, 64 J. POL. ECON. 416,
418 (1956) (presenting the classic model in which “[t]he consumer-voter may be
viewed as picking that community which best satisfies his preference pattern for public
goods[,]” and claiming that “[t]he greater the number of communities and the greater
the variance among them, the closer the consumer will come to fully realizing his preference position”). Regional migration patterns within the United States tend to show
that people do try to match their policy preferences to those of other citizens in choosing where to live. See Robert R. Preuhs, State Policy Components of Interstate Migration in
the United States, 52 POL. RES. Q. 527, 527-49 (1999) (finding that the “consumer-voter
model explains a significant portion of the variation in aggregate migration behavior”).
34
Representation in the United States Senate does substantially deviate from oneperson–one-vote principles. See SANFORD LEVINSON, OUR UNDEMOCRATIC CONSTITUTION: WHERE THE CONSTITUTION GOES WRONG (AND HOW WE THE PEOPLE CAN COR-
118
University of Pennsylvania Law Review
[Vol. 157: 103
These considerations do not mean that it never makes sense for a
polity to delegate authority upward or that doing so is necessarily undemocratic. They do show, however, that the resulting allocation of
authority will be somewhat less democratic than maintaining relatively
local control over the relevant policy question. Various reasons might
justify upward delegations of authority, including overcoming collective action problems, taking advantage of economies of scale, and coordinating regulation. Part III looks at these and other grounds for
dynamic incorporation in greater detail. My point here is simply that
the decision of a polity to be governed on some set of questions by the
decisions of others—including others of whom they are a proper subset—is almost always a decision for that polity to forego some measure
of self-government.
B. The Revocability Spectrum
Polity A’s decision to delegate decision-making power over some
class of questions to Polity B via dynamic incorporation (or via other
methods of delegation) does not divest A of sovereignty over this class
of questions, unless the delegation is irrevocable. Revocability, however, is not merely an on/off condition. It can be a matter of degree.
Ceteris paribus, the greater the formal and practical obstacles to revo-
RECT IT) 49-62 (2006) (critiquing the Constitution’s assignment of two senators to
each state, regardless of size, as fundamentally undemocratic). Accordingly, a decision
by a very small state like Vermont or Wyoming to dynamically incorporate federal law
does not dilute the political influence that individual Vermonters or Wyomingites exercise over the laws that govern them nearly as much as a similar decision by a large
state like California or Texas dilutes the political influence of Californians or Texans.
Nonetheless, even voters in the smallest states lose a measure of democratic representation when they cede authority to the federal government. According to the U.S.
Census Bureau, in 2006 the national population was just under 300 million. U.S. Census Bureau, State & County QuickFacts, http://quickfacts.census.gov/qfd/states/
00000.html (last visited Oct. 1, 2008). Thus, if the 100 Senate seats were apportioned
on a strictly proportional basis, it would take six million people to elect two senators.
In fact, however, the smallest state, Wyoming, with an estimated population of just over
half a million in 2006, id. at http://quickfacts.census.gov/qfd/states/56000.html,
elects two senators. Thus, Wyoming has roughly twelve times the representation in the
Senate as it would have if seats were allocated proportionate to population. However,
Wyomingites still only elect one-fiftieth of the U.S. Senate, whereas they elect all of the
Wyoming legislature. The twelvefold increase in influence relative to the national
population does not compensate for the fiftyfold decrease in influence that accompanies delegation to a much larger polity. Hence, even for population-challenged Wyoming, dynamic incorporation of federal law entails a democratic loss—one that is
compounded when we take account of the much-closer-to proportional influence that
Wyoming exercises over the choice of House members and the President.
2008]
Dynamic Incorporation
119
cation of a decision to dynamically incorporate the laws of a foreign
jurisdiction, the greater the threat that such dynamic incorporation
poses to democracy.
We can see the importance of revocability to democratic critiques
of dynamic incorporation by looking at the importance of revocability
to democracy more generally. Consider a leading example: Article
79(3) of the German Constitution purports to categorically forbid cer35
tain sorts of amendments, but scholars have questioned whether this
36
sort of permanent entrenchment is legitimate. What gave the drafters of the original Article 79(3) the right to decide any matter for all
time?
This is a hard question that also can be, and has been, asked about
the less severe forms of entrenchment that one commonly sees in constitutions. What gives any supermajority at any time the right to entrench its work against change by a mere ordinary majority? A satisfactory account of constitutional entrenchment—and thus of
constitutionalism itself—must provide an answer to this question.
We should distinguish two sorts of problems with entrenchment.
The first is simply a matter of positive law. Akhil Amar has argued that
the arduous amendment procedure set out in Article V of the U.S.
Constitution is not the exclusive means of amending the Constitution,
37
suggesting that a national referendum would also be effective. Like35
GRUNDGESETZ FÜR DIE BUNDESREPUBLIK DEUTSCHLAND [GG] [Federal Constitution] art. 79(3) (F.R.G.), translated in 7 CONSTITUTIONS OF THE COUNTRIES OF THE
WORLD 35 (Gisbert H. Flanz et al. eds., Supp. 2007) (“Amendments to this Basic Law
affecting the division of the Federation into Länder, their participation on principle in
the legislative process, or the principles laid down in Articles 1 and 20 shall be inadmissible.”).
36
See, e.g., Julian N. Eule, Temporal Limits on the Legislative Mandate: Entrenchment
and Retroactivity, 12 AM. B. FOUND. RES. J. 379, 386-96, 403-05 (1987) (examining objections to entrenchment in British and American contexts); Michael J. Klarman, Majoritarian Judicial Review: The Entrenchment Problem, 85 GEO. L.J. 491, 502-09 (1997) (describing majoritarian objections to legislative and cross-temporal entrenchment);
Frank I. Michelman, Entrenchment and the Dignity of Legislation 8 (Apr. 19, 2006)
(unpublished manuscript, on file with author) (arguing that the notion of entrenchment can only be gauged against a normative idea of representative democracy).
37
See Akhil Reed Amar, Philadelphia Revisited: Amending the Constitution Outside Article V, 55 U. CHI. L. REV. 1043, 1044 (1988) (“[T]he first, most undeniable inalienable
and important, if unenumerated, right of the People is the right of a majority of voters
to amend the Constitution—even in ways not expressly provided for by Article V.”); see
also Akhil Reed Amar, The Consent of the Governed: Constitutional Amendment Outside Article V, 94 COLUM. L. REV. 457, 457 (1994) (“[T]he U.S. Constitution is a far more majoritarian and populist document than we have generally thought; and We the People
of the United States have a legal right to alter our Government—to change our Constitution—via a majoritarian and populist mechanism akin to a national referendum,
120
University of Pennsylvania Law Review
[Vol. 157: 103
wise, one might think that Germans could circumvent Article 79(3)
simply by amending Article 79(3) itself. According to these sorts of
arguments, the U.S. and German Constitutions are best read—by
their own terms and in accordance with their respective histories—not
to be as deeply entrenched as they appear to be. These arguments
draw some support from claims about the democratic (il)legitimacy of
entrenchment, but they do so in order to better explain the scope of
entrenchment that actually exists under current law.
A second kind of problem with entrenchment concerns legitimacy
directly. Critics attentive to this problem typically begin with the (reasonable) assumption that contemporary majorities have the best claim
38
to legitimate lawmaking authority. According to this argument, one
generation never has the power to bind another, or if one generation
does have some power to bind later generations, the power must be
39
limited in some way. This sort of argument is sometimes used to at40
tack the practice of judicial review, but it also applies to constitutionalism itself.
I am not now interested in whether persuasive answers can be
given to critics of constitutional entrenchment. Instead, I merely wish
to note a common and apparently justified assumption in this debate:
as the degree of entrenchment of a constitutional provision (or its authoritative interpretation by a constitutional court) increases, so too
does the difficulty of reconciling the provision (or its interpretation)
with democratic principles. For example, a constitutional rule that
prohibits criminalization of abortion is more “countermajoritarian” in
a constitutional system like that of the United States, in which it can
even though that mechanism is not explicitly specified in Article V.”). But see Henry
Paul Monaghan, We the People[s], Original Understanding, and Constitutional Amendment,
96 COLUM. L. REV. 121, 121-22 (1996) (arguing that Amar “ignores the crucial role reserved for the states in the newly established constitutional order,” and that “the Constitution nowhere contemplates any form of direct, unmediated lawmaking or constitution-making by ‘the People’”).
38
See, e.g., Michael W. McConnell, Textualism and the Dead Hand of the Past, 66 GEO.
WASH. L. REV. 1127, 1127 (1998) (“The first question any advocate of constitutionalism
must answer is why Americans of today should be bound by the decisions of people
some 212 years ago.”).
39
See ERIC FONER, TOM PAINE AND REVOLUTIONARY AMERICA 216 (updated ed.
2005) (“Every age and generation must be as free to act for itself, in all cases, as the
ages and generations which preceded it.” (quoting THOMAS PAINE, The Rights of Man)).
40
See, e.g., Klarman, supra note 36, at 494-95 (critiquing judicial review on the
ground that standard sources of decision-making, including “[t]ext, original intent,
and tradition[,] are problematic . . . because of the dead hand problem”).
2008]
Dynamic Incorporation
121
41
only be reversed by a difficult process of constitutional amendment
42
(or through the judicial appointments process ), than it is in a consti43
tutional system like that of Canada, where it can be superseded by
the national Parliament acting pursuant to the “notwithstanding
44
clause” of the Charter of Rights and Freedoms.
To contend that the burden of persuasion on those who would
justify entrenchment increases as the degree of entrenchment increases is not to say whether or when that burden can be met. It is not
even to say that this is the only way of framing the problem. One
could, alternatively, assert that as the state seeks to regulate increasingly personal matters, its burden of justification also increases. One
could even argue that protection of certain fundamental freedoms is
so essential to constitutional democracy that the failure to protect
them—that is, the exposure of individuals to state regulation in the
area of these freedoms—would render the state less “democratic,”
45
even as it became more “majoritarian.” But this is a somewhat tendentious (which is not to say incorrect) use of the word “democratic.”
46
When I say that Roe v. Wade or, for that matter, the Equal Protection
47
Clause of the Fourteenth Amendment, is prima facie undemocratic,
all I mean is that the decision and the clause place limits on the laws
that duly elected legislative majorities can enact. Anyone who thinks
41
See U.S. CONST. art. V (requiring that either a two-thirds majority of both houses
of Congress or two-thirds of state legislatures must propose a constitutional amendment, and that either the legislatures or conventions of three-fourths of the states must
ratify an amendment before it is adopted).
42
See, e.g., Jack M. Balkin & Sanford Levinson, Understanding the Constitutional Revolution, 87 VA. L. REV. 1045, 1068 (2001) (suggesting that “[p]artisan entrenchment
through presidential appointments to the judiciary is the best account of how the
meaning of the Constitution changes over time through Article III interpretation
rather than through Article V amendment”).
43
The Canadian Supreme Court found criminal prohibition of abortion unconstitutional in R. v. Morgentaler, [1988] 1 S.C.R. 30.
44
Constitution Act, 1982, being Schedule B to the Canada Act 1982, ch. 11,
§ 33(1) (U.K.) (“Parliament or the legislature of a province may expressly declare in
an Act of Parliament or of the legislature, as the case may be, that the Act or a provision thereof shall operate notwithstanding [its violation of certain rights.]”).
45
See RONALD DWORKIN, FREEDOM’S LAW: THE MORAL READING OF THE AMERICAN
CONSTITUTION 17-18 (1996) (making this argument in support of a “constitutional,” as
opposed to a “majoritarian,” conception of democracy); JAMES E. FLEMING, SECURING
CONSTITUTIONAL DEMOCRACY: THE CASE OF AUTONOMY 19-36 (2006) (rejecting
methods of constitutional interpretation that seek only to perfect the processes of representative democracy).
46
410 U.S. 113 (1973).
47
U.S. CONST. amend. XIV, § 1.
University of Pennsylvania Law Review
122
[Vol. 157: 103
that I should use the word “majoritarian” rather than “democratic”
should feel free to substitute the former for the latter; such a disagreement on this point is semantic, not substantive.
This digression into constitutional law illustrates a commonplace
that verges on the tautological: from the standpoint of democratic
(or, if you prefer, majoritarian) values, entrenchment of constitutional provisions (or interpretations thereof) against simple majoritarian change stands in need of justification, and the greater the entrenchment, the greater the burden of justification.
We can readily generalize the point about entrenchment to other
limits on decision making placed upon officials elected within the
relevant polity: from the standpoint of democratic values, the greater
the entrenchment of the delegation to a body not directly accountable to the polity in question, the greater the burden of justification.
Other things being equal, it takes a stronger argument to justify Polity
A’s decision to dynamically incorporate Polity B’s law if the incorporating decision can only be undone by a supermajority vote of the legislature than it takes to justify an incorporating decision that can be
undone by a simple majority vote.
The fact that formal entrenchment makes dynamic incorporation
(and other sorts of delegation) more problematic should not obscure
the fact that even formally unentrenched incorporation raises prima
facie democratic problems because, as a practical matter, all legisla48
tion is entrenched in some sense.
Ordinary laws that can be repealed by simple-majority vote are actually formally entrenched (even
if only slightly), because obtaining a majority to repeal a law means
obtaining more votes than are required to prevent the enactment of
new legislation. Moreover, in legal systems like the United States and
all but one of its states, the requirement of bicameral passage and signature by the President or governor for “ordinary” legislation is, on its
49
face, a supermajority process. And, of course, as a practical matter,
the burden of overcoming legislative inertia even in a single house of
48
See Posner & Vermeule, supra note 27, at 1686-88 (“If there are political or logistical costs to repealing legislation—and there surely are—then an earlier Congress
‘binds’ a later Congress by enacting legislation that cannot be costlessly repealed or
changed, except in those instances when it provides for the legislation to expire on its
own.”).
49
See LEVINSON, supra note 34, at 29-49 (explaining how the bicameralism and
presentment requirements of the Constitution’s Article I, Section 7, protect the status
quo against new legislation).
2008]
Dynamic Incorporation
123
the legislature makes the repeal of legislation substantially more difficult than its nonenactment in the first place.
The closest thing to a completely unentrenched law is a law with a
sunset provision, but even these laws are entrenched during the period from their enactment until their sunset. Further, in some circumstances laws that formally sunset create expectations of renewal.
Most notoriously, appropriations measures that distribute concentrated benefits create powerful constituencies that lobby for their continual reenactment, as the U.S. experience with agricultural subsidies
50
demonstrates.
Despite the fact that such subsidies must be reen51
acted with each budget bill, critics of subsidies accurately complain
about the difficulty of “eliminating,” rather than “not reenacting,”
52
these technically sunsetting measures.
To be sure, dynamic incorporation by A of B’s law will not invariably create the sort of constituency for reenactment that agricultural
subsidies do, but in some circumstances the practical obstacles to
eliminating dynamic incorporation (regardless of whether the act of
incorporation formally sunsets) will be formidable. For instance,
when multiple polities collectively undertake to incorporate the future
enactments of a supranational body—as in the EU or NAFTA—and
when the only way to opt out of one of those enactments is to opt out
of the entire apparatus, with potentially disastrous economic and diplomatic consequences, we may treat the act of dynamic incorporation
as effectively irrevocable. Indeed, de facto irrevocability is often the
very point of such arrangements, precisely because supranational entities aim to provide assurances of reciprocal treatment, thus solving
collective action problems that could otherwise lead to suboptimal
53
outcomes such as trade wars.
*
50
*
*
See Dan Morgan et al., Powerful Interests Ally to Restructure Agriculture Subsidies,
WASH. POST, Dec. 22, 2006, at A1 (“Ever since subsidies began . . . Farm Belt politicians . . . have repeatedly thwarted efforts to scale [them] back . . . .”).
51
See Elizabeth Garrett, Rethinking the Structures of Decisionmaking in the Federal
Budget Process, 35 HARV. J. ON LEGIS. 387, 397-401 (1998) (explaining the budget process for the renewal of discretionary spending).
52
See, e.g., Morgan et al., supra note 50.
53
See Kenneth W. Abbott & Duncan Snidal, Hard and Soft Law in International Governance, 54 INT’L ORG. 421, 429 (2000) (describing the EU and NAFTA as “‘clubs’ of
sincerely committed states” purposely accepting the high costs of withdrawal in order
to “identif[y] . . . as having a low propensity to defect”).
University of Pennsylvania Law Review
124
[Vol. 157: 103
In sum, all instances of dynamic incorporation of foreign law raise
prima facie questions of democratic accountability. The entrenchment of the incorporated law against ordinary democratic processes
exacerbates the problem, but even when there is no formal entrenchment, the problem persists, in part because all law is entrenched in some sense.
II. DIRECTIONS OF INCORPORATION: UP, DOWN, AND ACROSS
This Part introduces a three-part typology of dynamic incorporation: upward incorporation, in which a political entity delegates lawmaking authority to a larger political entity; downward incorporation,
in which a political entity delegates lawmaking authority to a subunit;
and horizontal incorporation, in which a political unit dynamically incorporates the law of a comparable political unit.
A. Up
When a polity dynamically incorporates the law of a larger entity
of which it is a component, it incorporates upward. Within the United
States, state tax codes provide a familiar example. Some such codes
define “gross income” as the federal definition of gross income, even
if the federal definition changes (as it frequently does) after the en54
actment of the state incorporating law.
I shall also use the term upward incorporation to refer to decisions to dynamically incorporate some legal norms of supranational
bodies. When the Czech Republic joined the European Union, for
example, it agreed to make EU legal norms—including prospective
norms that the EU itself had not yet adopted at the time of the Czech
55
accession—applicable in the Czech Republic. This was upward incorporation even though, under the dominant Westphalian paradigm,
the Czech Republic did not cede any of its sovereignty to the EU (be56
cause accession is revocable). To be sure, under the German Solange
54
See supra note 14.
See Saulius Lukas Kaleda, Immediate Effect of Community Law in the New Member
States: Is there a Place for a Consistent Doctrine?, 10 EUR. L.J. [U.K.] l02, 102 (2004) (asserting that the principle that EU Community law becomes applicable in a member
state immediately upon accession nonetheless allows for some preexisting disputes to
be governed by the state’s preexisting law in accordance with “specific legal principles
[that] constitute part of community inter-temporal law”).
56
See Mattias Kumm, To be a European Citizen? The Absence of Constitutional Patriotism and the Constitutional Treaty, 11 COLUM. J. EUR. L. 481, 495 (2005) (arguing that,
55
2008]
Dynamic Incorporation
125
decisions and comparable rulings, European Constitutional Courts
purport to retain the authority to invalidate EU norms that violate na57
tional constitutions. But these rulings only show that an EU law that
is dynamically incorporated by an EU member state may sometimes
prove to be invalid, just as a domestic statute enacted by the parliament of an EU member state may be held invalid by that member
state’s constitutional court. EU accession counts as upward dynamic
incorporation because in ordinary cases, it gives the force of law to future EU laws and regulations that otherwise would have no effect
within the country in question.
B. Down
When a polity dynamically incorporates the law of one of its subunits, it incorporates downward. Within the United States, examples
of express decisions by Congress to dynamically incorporate the law of
58
states include the Assimilative Crimes Act, Federal Rule of Evidence
59
60
501, and the Rules of Decision Act —although in this last example
there is a colorable argument that Congress has no authority to make
federal law, and so state law applies of its own force by virtue of the
Tenth Amendment, rather than as a result of a congressional decision
61
to incorporate it dynamically. Putting this last caveat aside, in each
despite debate about the legal process for withdrawing from the EU, “the right to
withdrawal remain[s] a residual sovereign right, unencumbered by and antecedent to
EU Law”).
57
See Carlsen v. Prime Minister, [1998] 3 C.M.L.R. 854 (Den.) (holding that
European law that violates basic Danish constitutional rights is unenforceable in Danish courts); Solange II, supra note 9 (conditionally accepting European law’s supremacy
because it had developed sufficient fundamental-rights protection); Solange I, supra
note 9 (denying the supremacy of European law on the ground that European law
lacks fundamental-rights protection); Corte cost. [Constitutional Court], 13 apr. 1989,
n.232, Foro It. I, 1855 (Italy) (holding that European law must conform to the fundamental-rights protections of the Italian Constitution); see also supra note 9.
58
18 U.S.C. § 13 (2006); see also supra text accompanying note 19.
59
FED. R. EVID. 501; see also supra text accompanying note 18.
60
28 U.S.C. § 1652 (2006) (“The laws of the several states, except where the Constitution or treaties of the United States or Acts of Congress otherwise require or provide, shall be regarded as rules of decision in civil actions in the courts of the United
States, in cases where they apply.”).
61
This is the argument to which Justice Brandeis alludes in Erie Railroad Co. v.
Tompkins, 304 U.S. 64 (1938), suggesting that the Constitution requires the Rules of
Decision Act to be construed so as to deny federal courts the power to formulate general federal common law. At its core, Justice Brandeis’s argument asserts that where
Congress has no Article I power to regulate, it may not assign to the courts the power
to resolve disputes on the basis of judge-made rules. To be sure, given the expansive
126
University of Pennsylvania Law Review
[Vol. 157: 103
of these instances, Congress requires the application of state law, and
the question of which state’s law applies will be answered differently
for different cases.
We can also imagine a form of downward dynamic incorporation
that selects the law of a single subunit and makes it applicable as the
law of the whole larger entity. In particular industries, standard form
contracts sometimes come to include choice-of-law clauses that select
the law of a jurisdiction with no clear connection to the underlying
62
contract, and we can thus imagine that, pursuant to a mimic-themarket approach, some federal statute might do the same, at least as a
default rule. Consider, for example, a federal statute stating that federal contracts will be governed by the contract law of New York State.
63
Because much contract law is judge made, here the federal act of
dynamic incorporation would put into place future decisions of the
New York courts in addition to the New York legislature, but that
should not give us serious pause.
Indeed, true dynamic incorporation will typically incorporate not
only the future text of the statutes of the incorporated jurisdiction,
but also the future authoritative interpretations thereof. To give an
example in the upward incorporation context, states that require
“lockstep” interpretation of their state constitutional rights require
their state high courts to lock their steps with those of the United
States Supreme Court in its future interpretations of federal constitu64
tional rights.
modern view of the Commerce Clause—even after United States v. Lopez, 514 U.S. 549
(1995), and United States v. Morrison, 529 U.S. 598 (2000)—there are few subjects over
which Congress lacks Article I power. However, as George Rutherglen has argued persuasively, there is a further, horizontal, dimension to the constitutional view of Erie:
even where Congress has Article I power over some subject matter, it cannot delegate
that power to the courts on a wholesale basis but only in cases that happen to fall
within the courts’ jurisdiction under Article III and Title 28. See George Rutherglen,
Reconstructing Erie: A Comment on the Perils of Legal Positivism, 10 CONST. COMMENT. 285,
288, 295-96 (1993); see also Thomas W. Merrill, The Judicial Prerogative, 12 PACE L. REV.
327, 342-45 (1992) (challenging interpretations of Erie that “assum[e] federal courts
can exercise common law powers in any area that has been visited by congressional
legislation”).
62
See Jack M. Graves, Party Autonomy in Choice of Commercial Law: The Failure of Revised U.C.C. § 1-301 and a Proposal for Broader Reform, 36 SETON HALL L. REV. 59, 94-99
(2005) (surveying states that uphold contractual choice of law “irrespective of whether
the transaction has any relationship to the law chosen”).
63
See ROBERT E. SCOTT & JODY S. KRAUS, CONTRACT LAW AND THEORY 2 (rev. 3d
ed. 2003) (“Despite the statutory provisions that govern certain areas, much of modern
contract law is still based on common law decisions.”).
64
See supra note 16.
2008]
Dynamic Incorporation
127
C. Across
When a polity dynamically incorporates the law of a polity with
which it has no hierarchical relationship, it incorporates horizontally.
Examples of static horizontal incorporation abound. States attaining
independence frequently adopt “reception acts,” which keep the law
of their former colonial masters in effect pending the adoption of new
65
law. This amounts to horizontal incorporation at the point at which
the reception act becomes the law of the newly independent state,
because the new state stands on (newly) equal footing with the former
master state. However, reception acts usually operate statically. Postindependence changes in the law of the mother country do not, by
virtue of the reception act, automatically become the law of the
daughter state.
Examples of one especially unproblematic form of horizontal dynamic incorporation abound. States frequently authorize the application of the law of foreign states to transactions involving citizens or
subjects of that foreign state interacting with the host state. For example, California could be said to dynamically incorporate the law of
foreign states insofar as California (like other American states) permits residents of other American states and foreign nations who possess valid driver’s licenses from their states of residence to drive in
66
California. As the motor vehicle licensing laws of these other states
change, California’s treatment of out-of-state visitors changes along
with them. Indeed, California even permits unlicensed drivers to
65
See, e.g., MASS. CONST. pt. 2, ch. 6, art. 6 (“All the laws which have heretofore
been adopted . . . in . . . Massachusetts Bay . . . shall still remain and be in full force,
until altered or repealed by the legislature . . . .”); UNDANG -UNDANG DASAR REPUBLIK
INDONESIA 1945 [Constitution] Transitional Provisions, art. 1, translated in 9 CONSTITUTIONS OF THE COUNTRIES OF THE WORLD, supra note 35, at 19 (“All existing laws and
regulations shall remain in effect as long as new laws and regulations have not yet
taken effect under this Constitution.”). For an overview of reception acts in U.S. states,
see generally ELIZABETH GASPAR BROWN, BRITISH STATUTES IN AMERICAN LAW 1776–
1836, at 46 (1964), which summarizes each state’s reenactment of all or some British
statutes. On other former colonies retaining the law of their colonizers, see Ruth L.
Okediji, The International Relations of Intellectual Property: Narratives of Developing Country
Participation in the Global Intellectual Property System, 7 SING. J. INT’L & COMP. L. 315, 33435 & nn.73-74 (2003). See also M.B. HOOKER, LEGAL PLURALISM: AN INTRODUCTION
TO COLONIAL AND NEO-COLONIAL LAWS 360-409 (1975) (discussing “independent nation[s that have] of [their] own volition imported into [themselves] a legal system (or
parts of such a system) which is the product of a totally dissimilar legal culture”).
66
CAL. VEH. CODE §§ 12502–05 (West 2000); see also, e.g., FLA. STAT. ANN.
§ 322.04(c)-(d) (2005); N.Y. VEH. & TRAF. LAW § 250 (McKinney 2005); VA. CODE ANN.
§ 46.2-307 (2005).
University of Pennsylvania Law Review
128
[Vol. 157: 103
drive for a limited period if their state or country of residence does
67
not require licenses. Accordingly, whether a visitor to California may
drive without a license depends on whether her home state requires
licenses; should the foreign law change, her legal ability to drive in
California would change as well.
We can readily find other examples of this sort. Under familiar
conflicts principles, courts routinely apply the laws of foreign states,
either because the forum state’s conflicts law directly makes foreign
law applicable, or because the forum state’s contract law makes a vol68
untarily adopted choice-of-law provision enforceable. And unless the
parties to a contract otherwise specify, the operative law that will be
applied by the forum state will be the foreign law at the time that the
dispute arises, not at the time that the forum state (either legislatively
or by judicial decision) set forth the relevant choice-of-law or contract
rule.
State recognition of out-of-state marriages works in roughly the
same manner. Subject to a controversial public-policy exception, U.S.
states recognize marriages that took place in their sister states and
foreign countries, even when the couple would not have been eligible
69
to marry in the host state.
It is a nice semantic question whether to call laws of the foregoing
sort actual examples of dynamic horizontal incorporation. As a formal matter, it is true that state laws governing driving, conflicts, marriage, and some other subjects do make the legality of an act in State
A turn on the law of State B at some time that is at least potentially
later than the State A law went into effect. But these sorts of laws do
67
See CAL. VEH. CODE § 12503.
See RESTATEMENT (SECOND) OF CONFLICT OF LAWS § 6(1) (1971) (“A court, subject to constitutional restrictions, will follow a statutory directive of its own state on
choice of law.”); id. § 187(1) (“The law of the state chosen by the parties to govern
their contractual rights and duties will be applied if the particular issue is one which
the parties could have resolved by an explicit provision in their agreement directed to
that issue.”).
69
See, e.g., ARK. CODE ANN. § 9-11-107 (2008) (“All marriages contracted outside
this state that would be valid by the laws of the state or country in which the marriages
were consummated and in which the parties then actually resided shall be valid in all
the courts in this state. This section shall not apply to a marriage between persons of
the same sex.”); IDAHO CODE ANN. § 32-209 (2006) (“All marriages contracted without
this state, which would be valid by the laws of the state or country in which the same
were contracted, are valid in this state, unless they violate the public policy of this state.
Marriages that violate the public policy of this state include, but are not limited to,
same-sex marriages, and marriages entered into under the laws of another state or
country with the intent to evade the prohibitions of the marriage laws of this state.”).
68
2008]
Dynamic Incorporation
129
not raise the questions of democratic legitimacy that are the primary
concern of this Article—or at least do not raise them to the same degree as do other instances of dynamic incorporation.
When California permits a visiting Oregonian to drive because of
her compliance with Oregon law, California does (revocably) cede
70
some of its regulatory authority to Oregon, but it does so as a supplement to, rather than a replacement of, its own regulatory efforts.
Likewise, the decision to adjudicate a dispute based on the law of a
foreign jurisdiction typically rests upon either a voluntary commitment by the parties or the conclusion that the foreign jurisdiction has
a greater regulatory interest in the subject matter; the forum state continues to apply its own law to persons and transactions more closely
connected to the state. Accordingly, we would do better to focus on
horizontal circumstances in which Polity A makes the law of Polity B its
own law, rather than applying it to special cases as a supplement to the
law of A. Such cases are unusual, but they can be found.
Israel between 1948 and 1955 provides a case study. When Israel
attained independence from Great Britain in 1948, it continued to use
a colonial-era law that incorporated, on a dynamic basis, English com71
mon law decisions on questions as to which Israeli law was silent. By
1955, however, the Supreme Court of Israel realized that this practice
was anomalous. Thenceforth, the Court declared, it would treat Eng72
lish common law decisions as persuasive precedent only, although
70
Whether even this limited statement is true depends upon the extent to which
California is free, under the Federal Constitution, to prevent Oregon-licensed nonCalifornians from driving in California. The Dormant Commerce Clause, the Privileges and Immunities Clause of Article IV, and the Fourteenth Amendment all limit
that freedom. See Saenz v. Roe, 526 U.S. 489, 500-04 (1999) (identifying these constitutional provisions, as well as the constitutional structure itself, as sources of a constitutional right to travel); Bibb v. Navajo Freight Lines, 359 U.S. 520, 529-30 (1959) (holding that the Dormant Commerce Clause prohibits a state from enforcing laws that
“place a great burden of delay and inconvenience on . . . interstate motor carriers entering or crossing its territory”).
71
See Law and Administrations Ordinance, 5708-1948, 1 LSI 7 (1948) (Isr.)
(adopting, among other provisions, Article 46 of the Palestine Order in Council, 1922,
MOSES DOUKHAN, 2 LAWS OF PALESTINE 429, 431 (1933)); see also U. Yadin, Reception
and Rejection of English Law in Israel, 11 INT’L & COMP. L.Q. 59, 60-61 (1962) (noting
that, under the British Mandate, Israeli courts were to “rel[y] on English law [whenever the legal question] was a problem which fell into a gap, a lacuna, of the local
law”).
72
See Michael Zander, Book Review, 13 INT’L & COMP. L.Q. 1133, 1133 (1964) (reviewing E. DAVID GOITEN, Selected Judgements of the Supreme Court of Israel, 1948-1958
(1962)) (describing the decision in CA 92/55 Kohavi v. Becker [1955] P.D. 11 225).
University of Pennsylvania Law Review
130
[Vol. 157: 103
the Knesset did not formally repeal the act incorporating English
73
common law until 1980.
The view that dynamic horizontal incorporation is of dubious legitimacy appears to be widespread, and thus I have been able to find
relatively few examples in which one polity applies the law of another
unrelated polity as its own law. Nonetheless, there is a closely related
phenomenon that raises the same basic concerns. Consider East
Timor, Ecuador, El Salvador, the Federated States of Micronesia, the
Marshall Islands, Palau, and Panama. Each of these small, developing
countries has decided not to devote any of its limited resources to
managing its own monetary policy. Accordingly, none of these countries has its own currency, but instead has designated the United
74
States dollar as its official currency. Thus, U.S. laws and regulatory
actions governing the currency directly affect the currency of each of
these countries. Whenever the Federal Reserve Board, acting pursu75
ant to authority delegated by Congress, tightens or loosens the
money supply, its actions affect not only the United States currency
but, by virtue of the dynamic incorporation, also the money supply in
East Timor, even though East Timor is not in any way part of, or represented in, the United States.
In addition, some countries that maintain their own currency
“peg” that currency to the U.S. dollar, meaning that the government
76
commits to an official fixed exchange rate with the dollar. Some of
these pegs act simply as unofficial commitments on the part of foreign
governments to adopt fiscal and monetary policies that keep their cur77
rency at the preset exchange rate. However, where the decision to
peg is embodied in a law, that law might be said to dynamically incorporate U.S. law and regulatory actions governing the dollar, because
any subsequent changes in U.S. law or monetary policy—and con-
73
Foundations of Law, 5740-1980, 34 LSI 181 (1979–80) (Isr.).
See MONETARY AND CAPITAL MKTS. DEP’T, INT’L MONETARY FUND, REVIEW OF
EXCHANGE ARRANGEMENTS, RESTRICTIONS, AND CONTROLS 32 tbl.7 (Nov. 27, 2007),
available at http://www.imf.org/external/np/pp/2007/eng/112707.pdf (listing countries that have officially adopted or pegged to the currency of another country). Panama officially calls the U.S. dollars that it uses for currency “balboas,” even though
there are no actual balboa notes. See id. at 31 n.4.
75
See 12 U.S.C. § 411 (2006) (authorizing the Board of Governors of the Federal
Reserve system to issue federal reserve notes).
76
See PAUL R. KRUGMAN & MAURICE OBSTFELD, INTERNATIONAL ECONOMICS:
THEORY AND POLICY 447-58 (7th ed. 2005) (providing an overview of fixed exchange
rates).
77
See id. at 456-58.
74
2008]
Dynamic Incorporation
131
comitant effects of those changes on the dollar’s value against currencies that are not pegged to it—take effect automatically within the
pegging countries.
Needless to say, countries seeking a currency without creating one
of their own, or seeking stability through pegging, have choices besides dollars. Thus, for example, the official currency of Montenegro
is the euro, even though Montenegro is not part of the European Un78
ion. Likewise, since its introduction, other countries outside of the
79
EU have pegged their currency to the euro.
Whether a polity’s adoption of another sovereign’s currency or a
law formally pegging the polity’s own currency to a foreign currency
should be seen as dynamic incorporation of foreign law is not entirely
clear. The answer to that question might depend on what other laws
operate in the polity. For example, limits contained in laws governing
the interest rates that Panama banks can charge on loans, or pay on
deposits, could be triggered by the U.S. Federal Reserve Board’s decision to cut or raise interest rates, and if so, it would be fair to say that
the latter’s decision is dynamically incorporated through the Panama
law. But whether we call this or other variations on foreign currency
adoption and pegging pure or merely near examples of horizontal dynamic incorporation, this class of decisions will likely involve the same
sorts of costs and benefits as the more straightforward dynamic horizontal adoption of foreign regulatory norms that existed in Israel pre-1955.
As a practical matter, something reasonably close to dynamic horizontal incorporation of regulatory norms is quite common within the
United States. On questions of state corporate law to which corporate
codes do not spell out an answer in detail—matters such as the scope
of the business judgment rule, for example—many states treat Dela80
ware law as persuasive precedent. Although I have found no example of a state dynamically incorporating Delaware statutory or deci-
78
See MONETARY AND CAPITAL MKTS. DEP’T, supra note 74.
These include Eastern European countries, such as Estonia, Lithuania, and Latvia, id., as well as the fourteen African countries that use the CFA franc, which is itself
pegged to the euro, id. at 30 tbl.6.
80
See, e.g., Int’l Ins. Co. v. Johns, 874 F.2d 1447, 1459 n.22 (11th Cir. 1989) (“The
Florida courts have relied upon Delaware corporate law to establish their own corporate doctrines.”); Vogel v. Mo. Valley Steel, Inc., 625 P.2d 1123, 1126 (Kan. 1981)
(“The Kansas Corporation Code was patterned after the Delaware Corporation
Code . . . and therefore, Delaware decisions interpreting its code are considered persuasive in our interpretation of the Kansas code.”); McMinn v. MBF Operating Acquisition Co., 164 P.3d 41, 53 (N.M. 2007) (looking to Delaware corporate law for guidance, and citing cases from other states that do the same).
79
University of Pennsylvania Law Review
132
[Vol. 157: 103
sional law as an official matter, we could readily imagine a state that
hoped to replace Delaware as the corporate home for out-of-state corporations doing just this and undercutting Delaware on registration
81
fees. Indeed, it is something of a mystery why no state has tried do82
ing just that.
III. REASONS FOR DYNAMIC INCORPORATION
This Part describes and evaluates three of the most likely reasons
83
why a polity might find dynamic incorporation attractive. It considers three sorts of benefits that can flow from dynamic incorporation:
avoiding unnecessary costs by free-riding on the lawmaking efforts of
other polities; customizing the law to local conditions; and coordinating the efforts of actors in different jurisdictions. Each of these justifications comes with attendant costs as well as benefits, so that the availability of a particular argument for dynamic incorporation in some
context does not necessarily entail that dynamic incorporation is the
81
Corporate-law scholars frequently point to Nevada as an example of a jurisdiction that unsuccessfully attempted to supplant Delaware by copying its law. See Jill E.
Fisch, The Peculiar Role of the Delaware Courts in the Competition for Corporate Charters, 68 U.
CIN. L. REV. 1061, 1067 (2000) (“In addition to adopting the Delaware statute, the Nevada legislature adopted Delaware case law. Moreover, courts construing Nevada law
appear to follow Delaware precedent.”); Jonathan R. Macey & Geoffrey P. Miller, Toward an Interest-Group Theory of Delaware Corporate Law, 65 TEX. L. REV. 469, 488 (1987)
(“Nevada essentially has followed this course, adopting both the Delaware statutory
and common law as it applies to corporations.”); see also William L. Cary, Federalism and
Corporate Law: Reflections upon Delaware, 83 YALE L.J. 663, 665 (1974) (describing Nevada’s attempt “to become the western Delaware”). Whatever these scholars may think
Nevada did, the state did not, as a formal matter, dynamically incorporate Delaware
law. The authorities cited by the works listed in this footnote at most show that Nevada
statically copied provisions of the relevant Delaware statute and that Delaware cases are
persuasive precedent in Nevada. For example, Jill Fisch characterizes Hilton Hotels
Corp. v. ITT Corp., 978 F. Supp. 1342, 1346-47 (D. Nev. 1997), as “explicitly finding that
Nevada follows Delaware case law.” Fisch, supra, at 1067 n.45. In fact, this case only
states that Delaware law is persuasive authority where no Nevada statutory or case law is
on point. With respect to actual incorporation (of laws, not companies!), the most
that one can find in the Nevada case law is the statement that the Nevada statute, because it was based on a model act that was in turn rooted in Delaware and New York
case law, statically incorporates Delaware (and New York) case law. See Cohen v. Mirage
Resorts, Inc., 62 P.3d 720, 726 n.10 (Nev. 2003) (invoking the “rule of statutory interpretation that when a statute is derived from a sister state, it is presumedly adopted
with the construction given it by the highest court of the sister state” (internal quotation marks and citation omitted)).
82
Perhaps the best explanation is that the Delaware courts enjoy a reputational
advantage that an upstart rival would have great difficulty overcoming.
83
This Part does not argue that these are the only three reasons that might be offered in support of dynamic incorporation.
2008]
Dynamic Incorporation
133
best policy. Where relevant, this Part distinguishes between circumstances justifying upward, downward, and horizontal dynamic incorporation.
A. Avoiding Lawmaking Costs
Lawmaking is often costly. Where the institutional apparatus of
lawmaking already exists, many of the costs—such as salaries, benefits,
and expenses incurred by legislators and their staff, the cost of holding elections, and the building and maintenance of government office
buildings—are essentially fixed. The marginal cost of adding or
amending the legislative code is small along these dimensions.
However, legislation can have high marginal costs along other dimensions. Suppose the legislature in Polity A wants to regulate pollutants in drinking water. Formulating a legal rule or standard that
properly balances threats to human health against compliance costs
requires a mix of normative judgments and technical expertise. At a
minimum, either the legislature itself or an administrative agency to
which it delegates the task will need to undertake potentially expensive scientific studies. Saving these costs by writing law without adequate investigation risks imposing unwarranted back-end costs from
selecting a badly suboptimal rule.
Alternatively, the legislature in Polity A could free-ride on work
done in Polity B by simply adopting B’s rule or standard. This approach is especially attractive if the law requires continual updating in
light of new knowledge—as it plausibly might in an area like pollution
regulation—because dynamic incorporation permits A’s legislature to
make the one-time decision to depend on B’s law, and then to be
done with the matter.
To be sure, this strategy has drawbacks. The circumstances of A
may differ substantially from those of B. Perhaps A has a wetter climate than B, or B’s economy depends on water-intensive agricultural
production to a greater extent than does A’s. Or perhaps the people
in A simply value economic growth more (or less) relative to human
health than do the people of B. These sorts of differences will require
a dynamically (or, for that matter, a statically) incorporating polity
like A to be careful in selecting a jurisdiction B to ensure that B is sufficiently like A before borrowing its current and future law. Legislators in A will also want to consider whether the law of C might not be a
better choice than the law of B.
University of Pennsylvania Law Review
134
[Vol. 157: 103
Although these problems are real, they should not be overstated.
Even if Polity A cannot find another polity with the identical mix of
circumstances and values, the circumstances of some Polity B may be
close enough to those of A that the savings that accrue to A from not
having to create and continually amend its own substantive law more
than compensate for the defects, if any, in B’s law relative to the law
that A would ideally generate if left wholly to its own devices. That will
be especially true if A is a small polity and B a large one, as illustrated
by the fact that the countries that choose to use either the U.S. dollar
or the euro as their official currency, other than the countries that ac84
tually print the currency, tend to be quite small.
In principle, dynamic incorporation of foreign law as a cost-saving
device could proceed upwards, downwards, or horizontally. So long as
Polity A judges that the benefits of incorporating Polity B’s law outweigh the costs, it should not matter whether A is a subunit of B, B a
subunit of A, or neither is a subunit of the other. However, we are
unlikely to see downward dynamic incorporation purely on costsavings grounds unless the subunit whose law is incorporated by the
larger political entity has a greater capacity to generate optimal law
85
than most other subunits and the larger political entity itself.
When might that be true? Conditions favorable to downward dynamic incorporation on cost-savings grounds could be found in any
federal system in which a particular state government is thought to
have special expertise in a given subject area. In the United States, for
example, the Clean Air Act Amendments of 1977 provide for a version
of downward incorporation of California law, but only for those states
that wish to follow California law, and this incorporation is subject to
86
federal oversight.
84
See MONETARY AND CAPITAL MKTS. DEP’T, supra note 74, at 32 tbl.7.
I consider downward dynamic incorporation as a means of accommodating diversity below. See infra Part III.B (discussing circumstances where large political entities choose to devolve power to subunits in order to accommodate local preferences).
Such downward incorporation often will produce cost savings, but that is not its sole
justification.
86
See Clean Air Act Amendments of 1977, Pub. L. No. 95-95, § 209(b), 91 Stat. 755
(codified at 42 U.S.C. § 7543(b) (2000)) (permitting any state that regulates automobile emissions as strictly as the federal government to apply for a waiver, so long as the
applying state regulated such emissions prior to March 30, 1966). Because California
is the only state that did so, it alone qualifies for the waiver. However, the Clean Air
Act Amendments of 1977 permit other states to adopt the California standards and
thereby receive a waiver. Id. § 7507.
85
2008]
Dynamic Incorporation
135
In general, however, downward dynamic incorporation seems like
a sensible strategy in a federal system with both a weak and/or underfunded central government and one or a small number of powerful
state governments. Yet such federal systems will often be the product
of historical bargains between geographically based ethnic groups
with sufficient interethnic distrust that federal incorporation of the
87
law of a dominant state would likely be politically infeasible. For example, Belgian incorporation, and thus nationalization, of Flemish
law, or Czechoslovakian incorporation of Czech law during the brief
period between Czechoslovakia’s democratization and dissolution,
would be unacceptable to the minority ethnic groups (principally Wal88
loons and Slovaks, respectively).
Even static incorporation under
such circumstances would usually be problematic, and thus dynamic
incorporation, as a form of hegemony by one ethnic group over the
other(s), would almost certainly be regarded with suspicion.
Accordingly, we are most likely to see upward and horizontal dynamic incorporation, rather than downward incorporation, on laborsaving grounds. State tax codes that dynamically incorporate the federal definition of income (as well as other aspects of the Internal
89
Revenue Code) provide an example of double savings. First, by piggybacking on the federal definition, the state legislature saves itself
and its taxing authority the work of adjusting the law to changing circumstances. Second, this version of dynamic incorporation creates
cost savings for taxpayers. Rather than having to calculate their income (and possibly other terms prior to determining their tax liability) once for their federal forms and a second time for their state
forms, they can simply copy the result(s) from the federal form to the
state form. The time savings for state taxpayers are substantial, although less so now than a generation ago given the widespread avail87
See Donald L. Horowitz, The Many Uses of Federalism, 55 DRAKE L. REV. 953, 95758 (2007) (arguing that federalism generally arises in states plagued by ethnic conflict
only when potential or actual violence forces central governments to yield power to
subunits).
88
See generally JOHN FITZMAURICE, THE POLITICS OF BELGIUM: A UNIQUE FEDERALISM (1996) (noting that Belgium has developed a unique federal system that relies on a
system of complex institutions through which the various regions may discuss and
compromise on divisive issues rather than resort to violence); ERIC STEIN,
CZECHO/SLOVAKIA:
ETHNIC CONFLICT, CONSTITUTIONAL FISSURE, NEGOTIATED
BREAKUP (1997) (discussing the failed negotiations for a common state between
Czechs and Slovaks in 1992 and the role of ethnic conflict in shaping the outcome).
89
See supra note 14 (listing several state statutory and constitutional provisions that
incorporate definitions from the Federal Internal Revenue Code into state income tax
codes).
University of Pennsylvania Law Review
136
[Vol. 157: 103
ability of inexpensive computer software that can use the same raw
data to generate both federal and state returns.
As in our schematic water-pollution example, so too here the
benefits for state lawmakers and state citizens are not free. Changes in
federal tax law that become dynamically incorporated into state tax
law may result in revenue losses unwanted by state lawmakers or increased tax liability unwanted by state taxpayers. Nonetheless, the
90
widespread dynamic incorporation of federal tax law by state law
shows that many states regard these risks as cost justified.
As noted in Part II, true examples of horizontal dynamic incorporation are hard to find, but the examples that come closest—such as
the adoption by a small polity of a larger polity’s currency—will typically be justified on cost-savings grounds. Here, too, there is a tradeoff. In accepting U.S. monetary policy as its own, Palau sacrifices the
ability to expand or shrink its money supply in response to economic
91
conditions specific to Palau, but the cost savings for a nation with a
population that would just barely fill Madison Square Garden likely
outweigh this sacrifice.
Dynamic incorporation on cost-savings grounds will be especially
attractive where the polity whose law will be incorporated has, or is
perceived to have, special expertise in the relevant policy area. U.S.
states do not, as an official matter, dynamically incorporate the corporate law of Delaware, but they come close by treating Delaware corpo92
rate law decisions as especially persuasive precedent. Likewise, in
developing the federal common law of contracts that applies to gov93
ernment contracts, the federal courts may pay special attention to
the law of states (such as New York) that have the most extensively developed bodies of doctrine addressed to commercial matters. To be
clear, neither of these examples constitutes actual dynamic incorporation, as the target jurisdiction’s law is only deemed persuasive, and incorporation of Polity B’s judge-made law by the courts of Polity A
raises some questions not presented by legislative incorporation of
foreign legislation. Nonetheless, the examples demonstrate that expertise can be a reason to give some special place to the law of a for-
90
See supra note 14 and accompanying text.
Supra note 74 and accompanying text.
92
See supra notes 80-81 and accompanying text.
93
See Clearfield Trust Co. v. United States, 318 U.S. 363, 366-67 (1943) (holding
that federal common law governs federal contracts in suits by or against the federal
government).
91
2008]
Dynamic Incorporation
137
eign jurisdiction. Dynamic incorporation based on expertise simply
takes the point one (admittedly large) step further.
The arguments thus far considered for dynamic incorporation on
cost-savings and expertise grounds are all specific to particular policy
domains. However, the logic of cost savings may entail dissolution of
the polity entirely, and the fear of that logical consequence in turn
may lead polities to use upward dynamic incorporation less frequently
than simple cost-benefit analysis would counsel.
Consider Scotland. From the early eighteenth century through
1999, Scots voted in local elections and for ministers in the Parliament
94
of Great Britain, but had no Parliament of their own. In the debate
leading up to the 1997 referendum that resulted in a Scottish Parliament, the opposition emphasized cost (among other things). Running a government would be expensive, opponents argued, and the
money spent on the mechanics of a new layer of legislative government could be better spent on the provision of services or rebated
95
through lower taxes. That argument lost in the court of public opin96
ion, as the referendum passed by a three-to-one margin.
Now suppose that on some particular issue as to which formal
competence has been devolved to the Scottish Parliament, a Scottish
Member of Parliament (MP) argues that formulating the appropriate
legal standard would be too costly and that Great Britain as a whole is
similarly situated to Scotland. Accordingly, this MP suggests that the
Scottish Parliament should dynamically incorporate the British rule or
standard. We can readily imagine other MPs rising to argue that the
creation of the Scottish Parliament rules out this sort of “dedevolution.” Such an argument would rely on the very existence of
the polity’s own lawmaking bodies (which for these purposes include
administrative agencies to which responsibility for promulgating regulations with the force of law might be delegated) as a reason why the
substantive decisions must be made within the polity—even if decisions that are, by hypothesis, better would be reached by dynamic incorporation of the larger polity’s law.
94
See generally T.M. DEVINE, THE SCOTTISH NATION: A HISTORY, 1700-2000 (1999)
(recounting Scottish history from 1707 to the meeting of the first Scottish Parliament
in July of 1999).
95
See Sarah Lyall, With Gusto, Scots Say ‘Yes’ to Setting Up Parliament, N.Y. TIMES,
Sept. 12, 1997, at A3 (“Opponents of a new parliament in Scotland, in the ‘Think
Twice’ campaign led by the Conservative Party, have argued that it would saddle Scotland with new expenses [and] result in far higher taxes . . . .”).
96
Id.
University of Pennsylvania Law Review
138
[Vol. 157: 103
The argument that I have placed in the mouths of the hypothetical opponents of the hypothetical dynamic incorporation of British
law is available in any polity as an objection to upward incorporation.
“We Oklahomans have our own legislature for a reason,” the argument goes, “and thus even if Oklahoma would be better served by dynamically incorporating federal law, that would be an abdication of
our responsibility to legislate for the state.” As noted in the Introduction, in some U.S. states, this reasoning operates as a state constitu97
tional bar to dynamic incorporation, but even in states in which the
practice is not per se forbidden, the argument from the existence of
state lawmaking power may be invoked to resist particular efforts at
dynamic incorporation.
To be clear, I am not saying that those who invoke the existence
of lawmaking institutions as an argument against upward (or for that
matter, any) dynamic incorporation are correct. I am only noting that
the very existence of government institutions at any level of government will often tug against a decision to engage in dynamic incorporation. The loss—or even the perception of loss—of direct accountability for the law within the polity proper will count as a cost of
dynamic incorporation. For some, this cost will be seen as so high
that no benefit will justify incurring it, while for others it will merely
be thrown into the mix of overall costs and benefits. For anyone who
feels the tug of what might be called the “argument from the existence of government,” the argument counters claims that cost savings
justify dynamic incorporation as well as other sorts of justifications for
dynamic incorporation. The balance of this Part examines two such
other justifications for dynamic incorporation without separately raising the argument from the existence of government as an objection to
each, though each is vulnerable to that argument.
B. Accommodating Local Diversity Through Customization
Large political entities sometimes find it advantageous to devolve
power to their subunits. By permitting the law of the subunit to fill
gaps in or to entirely supply the law on particular issues, the larger
unit customizes the law to diverse local conditions and preferences.
Dynamic incorporation in this context permits the subunits to play the
97
See supra note 13 and accompanying text (enumerating twelve states that clearly
forbid dynamic incorporation of federal law as an unlawful delegation of lawmaking
authority, and three other states that forbid dynamic incorporation with limited exceptions).
2008]
Dynamic Incorporation
139
primary regulatory role in many substantive areas. Thus, even as devolution through dynamic incorporation decreases the reach of the larger polity, it increases the democratic character of the system as a
whole. In addition to matching policy to circumstances, devolution
also tends to encourage experimentation, which can in turn yield new
solutions that are then available for additional subunits or the polity as
a whole. This set of arguments for downward dynamic incorporation
closely parallels a familiar set of arguments for decentralization as a
98
more general matter.
As noted in the Introduction, within the United States there are
numerous examples of federal law rendering state law applicable
within the state or within some class of cases to which state law ap99
plies. These include the Conformity Act, which, prior to the adoption of the Federal Rules of Civil Procedure, required federal courts to
100
apply the procedural law of the states in which they respectively sat,
Federal Rule of Evidence 501, which makes state privilege law appli101
cable in federal court where state law provides the rule of decision,
and the Assimilative Crimes Act, which incorporates the criminal law
102
of the state in which federal land is located.
More broadly, Justice
Scalia was surely right in his observation that “there is nothing unusual about having the applicability of a federal statute depend on the
103
content of state law.”
Here, a digression on American constitutional doctrine will illuminate the broader phenomenon of downward dynamic incorporation. The constitutionality of downward dynamic incorporation by the
federal government was not firmly established until 1957, when the
Supreme Court upheld the Assimilative Crimes Act in United States v.
98
See Michael C. Dorf & Charles F. Sabel, A Constitution of Democratic Experimentalism, 98 COLUM. L. REV. 267, 283-89 (1998) (arguing that a model of decentralized government allows localities to address particular concerns while learning from the experience of their neighbors, encourages experimentation and reform, and enables
citizens to directly participate in service provision and to evaluate the effectiveness of
local government institutions).
99
See supra text accompanying notes 17-19 (providing examples of federal law dynamically incorporating state law).
100
Supra note 17 and accompanying text.
101
FED. R. EVID. 501 (“[T]he privilege of a witness, person, government, State, or
political subdivision thereof shall be determined in accordance with State law.”).
102
18 U.S.C. § 13 (2006).
103
Stewart Org. v. Ricoh Corp., 487 U.S. 22, 35 n.* (1988) (Scalia, J., dissenting)
(internal quotation marks omitted) (noting Comm’r v. Estate of Bosch, 387 U.S. 456,
457, 464-465 (1967) and Budinich v. Becton Dickinson & Co., 486 U.S. 196, 199
(1988) as prior examples).
University of Pennsylvania Law Review
140
[Vol. 157: 103
104
105
Sharpnack.
Even then, neither the Court nor the appellee’s brief
cited the best authority for the proposition that federal law cannot dynamically incorporate state law.
In Cooley v. Board of Wardens, Justice Curtis, speaking for the Court,
upheld an 1803 Pennsylvania law that required the use of local pilots,
thereby refusing to rely on a 1789 federal statute purporting to au106
thorize state and local harbor laws.
Justice Curtis made clear that
Congress could have statically incorporated state law, but asserted
that—at least if a power were exclusively federal—the federal government could not, in the exercise of that power, dynamically incorpo107
rate state law.
He concluded: “If the Constitution excluded the
States from making any law regulating commerce, certainly Congress
cannot regrant, or in any manner reconvey to the States that
108
power.”
To be sure, this was dicta, because the Court held in Cooley that
the Constitution did not “exclude[] the States from making any law
109
regulating commerce.” The case was important for establishing the
proposition that, with the exception of the ill-defined category of objects of regulation that “are in their nature national, or admit only of
110
one uniform system, or plan of regulation,” states have concurrent
power to regulate commerce, unless and until Congress preempts
their regulations. And because Cooley upheld what amounted to a
devolution of power from Congress to the states in effect not unlike
104
105
106
107
355 U.S. 286 (1958).
Brief for Appellee, Sharpnack, 355 U.S. 286 (No. 35), 1957 WL 87702.
53 U.S. (12 How.) 299 (1851).
The Court stated,
If the law of Pennsylvania, now in question, had been in existence at the
date of this act of Congress, we might hold it to have been adopted by Congress, and thus made a law of the United States, and so valid. Because this act
does, in effect, give the force of an act of Congress, to the then existing State
laws on this subject, so long as they should continue unrepealed by the State
which enacted them.
But the law on which these actions are founded was not enacted till
1803. . . .
If the States were divested of the power to legislate on this subject by the
grant of the commercial power to Congress, it is plain this act could not confer upon them power thus to legislate.
Id. at 317-18.
108
Id. at 318.
109
Id.
110
Id. at 319.
Dynamic Incorporation
2008]
141
the devolution accomplished by the Assimilative Crimes Act, it is not
really surprising that, in the days before computer-based legal research, the appellee’s lawyer in Sharpnack (apparently a solo practitioner arguing his only Supreme Court case) did not think to look to
Cooley for support.
But we have no such excuse, and so we should face the question of
whether Justice Curtis and the Cooley Court were right in 1851 or
whether the Sharpnack Court was right over a century later: does the
Constitution permit Congress to dynamically incorporate downward
when it exercises an exclusive federal power (as Congress does pursuant to the Territories Clause of Article IV, Section 3, when, as under
the Assimilative Crimes Act, it makes law for federal territories)?
History and precedent would appear to be on the side of the
Sharpnack Court. In addition to the passage by the first Congress of
the federal act at issue in Cooley—which, on its face, incorporates state
111
law dynamically rather than just statically —the Conformity Act of
1872 (admittedly adopted after the Cooley decision) expressly displaced a regime of static conformity to state procedure under the
112
113
Process Act of 1792 with a regime of “dynamic conformity.”
Moreover, as the Sharpnack Court itself noted, numerous other federal
114
statutes in force by 1957 dynamically incorporated state law.
Did Justice Curtis nonetheless have logic on his side? Consider a
Constitution like that of Germany, which divides federal subject matter competencies into two categories: those over which the federal
115
government exercises exclusive authority and those over which the
111
The Act provided:
[t]hat all pilots in the bays, inlets, rivers, harbors, and ports of the United
States shall continue to be regulated in conformity with the existing laws of
the States, respectively, wherein such pilots may be, or with such laws as the
States may respectively hereafter enact for the purpose, until further legislative provision shall be made by Congress.
Id. at 317 (emphasis added) (quoting Act of Aug. 7, 1789, ch. 9, § 4, 1 Stat. 53, 54).
112
Act of May 8, 1792, 1 Stat. 275, 276; see also RICHARD H. FALLON, JR. ET AL.,
HART AND WECHSLER’S THE FEDERAL COURTS AND THE FEDERAL SYSTEM 605-07 (5th
ed. 2003) (noting that the Process Act “adopted the state law ‘as it existed in September, 1789 . . . not as it might afterwards be made’” (quoting Wayman v. Southard, 23
U.S. (10 Wheat.) 1, 32 (1825))).
113
See FALLON ET AL., supra note 112, at 607 (“The Conformity Act eliminated the
anachronism of federal adherence to no-longer-existent state practice.”).
114
United States v. Sharpnack, 355 U.S. 286, 294-97 (1958).
115
See GRUNDGESETZ FÜR DIE BUNDESREPUBLIK DEUTSCHLAND [GG] [Federal Constitution] art. 73, translated in 7 CONSTITUTIONS OF THE COUNTRIES OF THE WORLD,
142
University of Pennsylvania Law Review
[Vol. 157: 103
116
Länder and the federal government exercise concurrent authority.
Now suppose that, within an area of exclusive federal competence, the
national legislature enacts a law purporting to dynamically incorporate Länder law. This certainly looks like an effort to reclassify an exclusively federal power as a concurrent power. Thus it would likely be
unconstitutional were it not for another provision of the German
Constitution, which expressly empowers the federal government to
delegate lawmaking authority to the Länder even with respect to sub117
jects within exclusive federal jurisdiction.
Were it not for that additional provision, however, it would be a fair inference that exclusive
federal powers are not delegable to the Länder.
Likewise, where the U.S. Constitution clearly delineates a power as
exclusively federal, dynamic incorporation of state law would appear
to be forbidden. Only one provision of the Constitution expressly
confers an “exclusive” power. Article I authorizes Congress to “exercise exclusive Legislation in all Cases whatsoever” with respect to the
118
capital district. Nonetheless, Congress has granted (and taken away
and regranted) the District of Columbia varying degrees of “home
119
In 1953, the Supreme Court
rule” over the course of its existence.
rejected the argument that such home rule violates the constitutional
requirement of exclusively federal legislation, observing “that the
word ‘exclusive’ was employed to eliminate any possibility that the legislative power of Congress over the District was to be concurrent with
120
that of the ceding states.”
Thus, the Constitution’s grant of “exclu-
supra note 35, at 27 (enumerating areas in which the federal government has “exclusive power to legislate”).
116
See id. art. 74, translated in 7 CONSTITUTIONS OF THE COUNTRIES OF THE WORLD,
supra note 35, at 29 (extending “[c]oncurrent legislative powers” to enumerated subjects).
117
Article 71 of the German Constitution provides that “[o]n matters within the
exclusive legislative powers of the Federation, the Länder shall have the power to legislate only when and to the extent that they are expressly authorized to do so by a federal law.” Id. art. 71, translated in 7 CONSTITUTIONS OF THE COUNTRIES OF THE WORLD,
supra note 35, at 26.
118
U.S. CONST. art. I, § 8, cl. 17.
119
For a brief history of the relationship between Washington, D.C. and the federal government, see Note, Democracy or Distrust? Restoring Home Rule for the District of
Columbia in the Post–Control Board Era, 111 HARV. L. REV. 2045, 2046-51 (1998).
120
District of Columbia v. John R. Thompson Co., 346 U.S. 100, 109 (1953) (citing
THE FEDERALIST NO. 43 ( James Madison); 3 DEBATES 432-433 ( Jonathan Elliot ed., 2d
ed. 1897); 2 JOSEPH STORY, COMMENTARIES ON THE CONSTITUTION OF THE UNITED
STATES § 1218 (4th ed. 1873)).
2008]
Dynamic Incorporation
143
sive” power is not delegable to the states in this context, but mostly
because no state has a clear interest.
Naturalization and bankruptcy provide additional examples of potentially nondelegable powers. Article I empowers Congress “[t]o establish a uniform Rule of Naturalization, and uniform Laws on the
121
subject of Bankruptcies throughout the United States.”
A federal
naturalization or bankruptcy law that made the law of each state operative in its own territory would seem to violate the requirement of
uniformity. Note, however, that even static federal incorporation of
the law of fifty different states would violate uniformity on this theory.
Conversely, a federal law that dynamically incorporated the law of a
single state and made it applicable to the nation as a whole would not
be objectionable on disuniformity grounds. Accordingly, these uniformity requirements do not provide a very good test of the question
of whether exclusive federal power precludes dynamic, territorially
based incorporation of state law.
In fact, it turns out that federal bankruptcy law does dynamically
incorporate the law of the several states, thus seeming to violate the
uniformity requirement. Although the Bankruptcy Code begins by
122
defining terms, even a casual perusal of those terms makes obvious
that these terms are, in turn, defined by reference to relationships
that the Federal Code does not itself define but that take their substance from state law. For example, although the Code defines a
“domestic support obligation,” that definition relies on the underlying
123
law of domestic relations.
In principle, Congress, in enacting the
Bankruptcy Code, could have intended for the federal courts to develop a freestanding federal law of domestic relations applicable in
bankruptcy cases—and the Supreme Court has sometimes asserted
that there is a presumption in favor of federal definitions of federal
124
statutory terms.
But such statements cannot be taken at face value.
Although the development of federal standards in principle promotes
the uniformity of federal law, where, as in the Bankruptcy Code and
elsewhere, federal law assumes an existing set of legal relationships,
121
U.S. CONST. art. I, § 8, cl. 4.
See 11 U.S.C. § 101 (2006).
123
See id. § 101(14A).
124
See Jerome v. United States, 318 U.S. 101, 104 (1943) (“Congress when it enacts
a statute is not making the application of the federal act dependent on state law.”); see
also Miss. Band of Choctaw Indians v. Holyfield, 490 U.S. 30, 43 (1989) (starting with
the general assumption that the legislature does not intend for a statutory term of a
federal act to be given content by the application of state law).
122
University of Pennsylvania Law Review
144
[Vol. 157: 103
federal uniformity comes at the unacceptably high cost of almost random unpredictability. Must federal bankruptcy courts develop federal
standards governing alimony, divorce, guardianship, and separation
(to name just four legal relations and obligations to which one of over
fifty definitional sections of the Bankruptcy Code refers)? And to
what end? Whatever the merits of the outcomes reached in the cases
in which the Supreme Court has broadly (and blithely) asserted the
presumption in favor of federal definitions, surely the Justices were
closer to the mark when they said,
The scope of a federal right is, of course, a federal question, but that
does not mean that its content is not to be determined by state, rather
than federal law. This is especially true where a statute deals with a familial relationship; there is no federal law of domestic relations, which is
125
primarily a matter of state concern.
To be clear, I am making a normative claim that federal law should
not be interpreted so as to require the formulation of federal definitions for all of the legal relationships, rights, and duties to which it re126
fers, but I am also making a descriptive claim: federal law frequently
builds on state law as it finds it at the moment of application. In other
words, in many instances federal law dynamically incorporates state
law. And that is true even of nominally uniform federal law—such as
the federal law of bankruptcy.
Could it be otherwise? Once we recognize that in many instances
federal law will incorporate state law rather than supply all of its own
terms, can we say with confidence that it does so dynamically rather
than statically? We can. Despite the enormous growth of the federal
government since its founding, Herbert Wechsler’s 1954 observation
remains largely true even in 2008: “[F]ederal law is still a largely interstitial product, rarely occupying any field completely, building nor127
mally upon legal relationships established by the states.”
When
Wechsler said, accurately, that federal law builds on these state-
125
De Sylva v. Ballentine, 351 U.S. 570, 580 (1956) (citation omitted).
For an instructive discussion of the materials relevant to the normative question, see FALLON ET AL., supra note 112, at 723-24.
127
Herbert Wechsler, The Political Safeguards of Federalism: The Role of the States in the
Composition and Selection of the National Government, 54 COLUM. L. REV. 543, 545 (1954).
But see FALLON ET AL., supra note 112, at 495 (observing that “today one finds many
more instances in which federal enactments supply both right and remedy in, or
wholly occupy, a particular field,” but wondering whether Wechsler’s thesis, which was
also espoused by Henry Hart, “does not remain accurate over an extremely broad
range of applications”).
126
2008]
Dynamic Incorporation
145
defined legal relationships—including such fundamental matters as
128
129
130
contracts, family relations, and property —what he meant was
that federal law dynamically incorporates state law.
Incorporation of state law in such circumstances must be dynamic
in order for the incorporation to achieve one of its central purposes:
to layer federal law on top of the operative state law in any given case.
Static incorporation would require keeping track of the operative dates
of federal law and the law of each state, with incongruous results. It
would require, among other things, a determination of which amend131
ments to federal law reset the clock for purposes of incorporation.
Dynamic incorporation also makes more sense than static incorporation as a matter of congressional intent. Consider a tax example:
federal law treats punitive, but not most compensatory, damages for
132
personal injuries as part of gross income.
Subject to federal constitutional limits, state law defines when punitive damages are allowed
and where the line lies between punitive damages and compensatory
damages. To some extent, the distinction is a matter of labeling because juries may base their awards on a gestalt reaction to the facts of
the case, rather than a careful parsing of the differences among economic damages, compensatory noneconomic damages, and punitive
128
See, e.g., First Options of Chicago, Inc. v. Kaplan, 514 U.S. 938, 944 (1995)
(“When deciding whether the parties agreed to arbitrate a certain matter [under the
Federal Arbitration Act], courts generally . . . should apply ordinary state-law principles
that govern the formation of contracts.”).
129
See, e.g., Califano v. Jobst, 434 U.S. 47, 52-53 n.8 (1977) (enumerating several
provisions of the Social Security Act that incorporate marital status as defined by state
law); Gillett-Netting v. Barnhart, 371 F.3d 593, 599 (9th Cir. 2004) (conditioning entitlement to insurance benefits on state law determinations of whether children are legitimate); Purganan v. Schweiker, 665 F.2d 269, 270-71 (9th Cir. 1982) (holding that a
California law on marriage annulment affects insurance benefits under the Social Security Act).
130
See, e.g., United States v. Craft, 535 U.S. 274, 278 (2002) (“The federal tax lien
statute itself ‘creates no property rights but merely attaches consequences, federally
defined, to rights created under state law.’” (quoting United States v. Bess, 357 U.S. 51,
55 (1958))).
131
For an excellent discussion of this difficulty in the context of a purportedly
nonsubstantive change to the Federal Rules of Civil Procedure, see Edward A. Hartnett, Against (Mere) Restyling, 82 NOTRE DAME L. REV. 155 (2006).
132
I.R.C. § 104(a)(2) (2006). The evident object of the distinction is clear. Because the law does not tax imputed income from healthy bodies, damages for the loss
of healthy bodies should not be treated as income. Note, however, that the law also
does not tax imputed income from emotional well-being, but damages for emotional
distress are only partially exempt from gross income. See id. § 104(a).
University of Pennsylvania Law Review
146
[Vol. 157: 103
133
damages. Nonetheless, these labels matter for federal tax purposes.
But is it reasonable to suppose that, when Congress chose to accord
significance to the state-drawn line between punitive and most compensatory damages, it meant to adopt the lines as they existed in each
of the fifty states at the moment of enactment of the federal law—an
administrative nightmare—or the line as it exists in particular verdicts
based on the state law as it stands at the moment of that verdict, even
if that state law became operative after the relevant provision of the
federal tax law?
The pervasiveness of federal dynamic incorporation of state law
shows that Justice Curtis was not merely wrong in his Cooley dictum
denying the federal government this power, but very wrong. Or if he
was right, he was right only as to the almost vanishingly small category
of truly exclusive federal powers. A federal law that dynamically incorporated state law with respect to some aspect of foreign affairs—
“incorporating” Massachusetts law regarding transactions with firms
134
conducting business in Burma/Myanmar, say—might violate the
principle that in the exercise of its truly exclusive federal powers,
Congress cannot dynamically incorporate state law. But such exotic
examples aside, a federal system in which federal law is interstitial in
the way that Wechsler described cannot operate effectively without
dynamically incorporating state law willy-nilly.
C. Coordination and Reciprocity
Dynamic incorporation of foreign law can be a powerful mechanism for solving coordination problems and ensuring reciprocal compliance with agreements among sovereigns or quasi-sovereigns. By
contrast with static incorporation, dynamic incorporation ensures that
initial efforts at coordination do not drift apart over time.
Consider, first, a common alternative mechanism for achieving
coordination among various jurisdictions: the adoption by each of a
“uniform” or “model” code. The Uniform Commercial Code (UCC)
is a leading example. Originally promulgated in 1952 and revised periodically since then, some version of the UCC is in force in all fifty
133
Cf. Catherine M. Sharkey, Unintended Consequences of Medical Malpractice Damages
Caps, 80 N.Y.U. L. REV. 391 (2005) (finding that caps on noneconomic damages lead
to higher awards for economic damages).
134
Cf. Crosby v. Nat’l Foreign Trade Council, 530 U.S. 363, 366-67 (2000) (finding
that federal law preempted a Massachusetts law that prohibited that state’s agencies
from doing business in Burma).
Dynamic Incorporation
2008]
147
135
states.
States adopting the UCC may do so for one or both of two
principal reasons: First, the legislature in a given state may conclude
that the American Law Institute and the National Conference of
Commissioners on Uniform State Laws—the bodies responsible for
writing and updating the UCC (and other uniform laws)—have
greater expertise than legislators in drafting appropriate laws to govern commercial affairs. Second, given the large number of commercial transactions that cross state lines, states may conclude that uniform rules are preferable to custom-tailored rules. Even if State X
could enact a code that was superior to the UCC, the State X legislature might nonetheless conclude that the advantages of a set of uniform rules outweigh the disadvantages arising out of the (assumed)
suboptimality of the UCC rules. Accordingly, many states simply
adopt the UCC or individual articles thereof.
However, not every state adopts every article of the UCC, and
states sometimes adopt parts of the UCC with changes chosen by their
136
individual legislatures. Moreover, even if every state has adopted an
137
article of the UCC, individual states do so statically. How to interpret a provision of the UCC in Wisconsin is a question of Wisconsin
law, while how to interpret the identical language in Minnesota is a
question of Minnesota law. Ceteris paribus, in order to maintain uniformity, state courts may prefer interpretations adopted by the courts
135
See U.C.C. Table of Jurisdictions Wherein Code Has Been Adopted, 1 U.L.A. 12 (2004).
136
For example, U.C.C. § 2-302 (2007) sets forth a standard for an unconscionable
contract or clause. Forty-eight states have adopted section 2-302 in order to regulate
the perceived fairness of contracts. Ronald L. Hersbergen, Unconscionability: The Approach of the Louisiana Civil Code, 43 LA. L. REV. 1315, 1315, 1317 (1983). California has
not adopted the UCC’s unconscionability provision, but instead applies its own “judicially imposed” unconscionability doctrine. See, e.g., Graham v. Scissor-Tail, Inc., 623
P.2d 165, 172-78 (Cal. 1981) (per curiam) (observing that a contract may be adjudged
unconscionable and deemed unenforceable if it does not fall within the reasonable
expectations of the weaker party or if it is unduly oppressive). Louisiana, on the other
hand, does not expressly impose any unconscionability limit. See, e.g., Iberia Credit
Bureau, Inc. v. Cingular Wireless LLC, 379 F.3d 159, 167 (5th Cir. 2004) (“No section
of the Louisiana Civil Code directly addresses, in so many words, the doctrine of unconscionability or the related concept of adhesionary contracts.”); La. Power & Light
Co. v. Mecom, 357 So. 2d 596, 598 (La. Ct. App. 1978) (“The Louisiana courts have
not to date adopted the theory of adhesion contracts.”).
137
According to the website of the Uniform Law Commissioners, every state (as
well as the District of Columbia and the U.S. Virgin Islands) has adopted Article 9 of
the UCC, governing secured transactions. See Uniform Law Comm’rs, A Few Facts
about the Revised UCC Article 9, Secured Transactions (1999) (2002),
http://www.nccusl.org/Update/uniformact_factsheets/uniformacts-fs-ucca9.asp (last
visited Oct. 1, 2008).
University of Pennsylvania Law Review
148
[Vol. 157: 103
of their sister states, but this is at best an imperfect strategy for maintaining uniformity. Different fact patterns will arise in different orders in different states, and thus lines of doctrine interpreting uniform but somewhat ambiguous language may diverge over time. More
138
fundamentally, since Erie Railroad Co. v. Tompkins, American lawyers
and judges have understood state law (whether common law or statutory) as validating distinctive state authority rather than some general
law. Despite its name, the Uniform Commercial Code can never be
fully uniform.
If respectful consideration of the persuasive precedents of foreign
jurisdictions is an imperfect method of achieving harmonization, dynamic incorporation marks an improvement. Suppose that instead of
adopting the Uniform Commercial Code, Minnesota were to dynamically incorporate Wisconsin’s version of the Uniform Commercial
Code, along with its authoritative interpretation by the Wisconsin
courts. Now there would be a self-correcting limit to drift. On questions of first impression, the Minnesota courts would either interpret
the UCC according to their own best judgment or, following the practice of federal courts when called upon to make rulings of state law,
attempt to predict how the Wisconsin Supreme Court would rule on
139
the question. But once the Wisconsin Supreme Court actually ruled
on some question arising under the UCC, that ruling would be binding in Minnesota. The commercial law would remain uniform between Wisconsin and Minnesota.
In practice, it is difficult to imagine one state delegating so much
authority to another state in this manner, and harder still to imagine
forty-nine states delegating such authority to just one—although the
persuasive weight given to Delaware corporate law comes fairly close
in practice. As noted above, setting aside special cases involving the
application of external law to travelers and conflicts of law, true instances of horizontal dynamic incorporation are difficult to find.
More commonly, polities looking for a mechanism to keep their
law synchronized will dynamically incorporate upward the law of some
larger political entity. Multilateral treaties can serve this purpose.
Each member state signs onto the treaty and makes the law of the larger entity part of its own law—or, what amounts to much the same
138
304 U.S. 64 (1938).
See Salve Regina College v. Russell, 499 U.S. 225, 241 (1991) (Rehnquist, C.J.,
dissenting, but not on this point) (“[W]here the state law is unsettled . . . the courts’
task is to try to predict how the highest court of that State would decide the question.”).
139
2008]
Dynamic Incorporation
149
thing, makes the law of the larger entity superior to its own law and
enforceable in its own courts.
We might regard the U.S. Constitution itself as a leading example
of this sort of upward dynamic incorporation for the purpose of solving coordination problems and ensuring reciprocal compliance with
mutually beneficial agreements among sovereigns or quasi-sovereigns.
The states granted to the national government the power, among
other things, to regulate interstate commerce. And through the Supremacy Clause, actions of the federal government in this and other
domains of its policy competence were made both superior to con140
trary state law and enforceable within state courts. Thus, the Constitution gave the states a mechanism to achieve uniformity in interstate
commercial regulation and to avoid costly trade wars.
The example of the U.S. Constitution also highlights an important
feature of dynamic incorporation. Where there is a real possibility
that issues of interpretation will arise over time, in order for the original act of incorporation to count as truly dynamic, there must be some
mechanism for coordinating subsequent interpretations. Typically,
that mechanism will be submission of contested questions to a single
tribunal whose rulings will be binding on the courts of the incorporating polities. In the early nineteenth century, the highest court of Virginia took the position that its interpretation of the federal Constitution was no less authoritative than the interpretation provided by the
Supreme Court of the United States. Unsurprisingly, the latter Court
concluded that its own interpretation was binding, principally relying
on an argument rooted in the need for uniform interpretation of fed141
eral law.
Many years later, Justice Holmes famously remarked that
142
this ruling was essential to the preservation of the Union.
Thus, we might conclude that, in general, parties entering a multilateral agreement aimed at overcoming coordination problems should
also establish some body to issue authoritative decisions resolving
questions arising out of ambiguities in the agreement. This aim might
140
See U.S. CONST. art. VI, cl. 2 (“This Constitution, and the Laws of the United
States which shall be made in Pursuance thereof . . . shall be the supreme Law of the
Land; and the Judges in every State shall be bound thereby, any Thing in the Constitution or Laws of any state to the Contrary notwithstanding.”).
141
See Martin v. Hunter’s Lessee, 14 U.S. (1 Wheat.) 304, 347-48 (1816) (invoking
the “necessity of uniformity of decisions throughout the whole United States”).
142
See OLIVER WENDELL HOLMES, Law and the Court, in COLLECTED LEGAL PAPERS
291, 295-96 (1920) (“I do not think the United States would come to an end if we lost
our power to declare an Act of Congress void. I do think the Union would be imperiled if we could not make that declaration as to the laws of the several States.”).
University of Pennsylvania Law Review
150
[Vol. 157: 103
be accomplished by the creation of a legislature, an administrative
body, a court, or some combination. But whatever the nature of the
body or bodies, they will only confer on the parties the benefits of true
dynamic incorporation if their pronouncements are effectively binding within the legal systems of the agreeing parties. At the same time,
however, bestowing the power to make binding rules of law on a body
that is in substantial part beyond the control of the polity granting the
power at issue renders dynamic incorporation potentially problematic
on delegation or sovereignty grounds.
*
*
*
The foregoing catalogue of the sorts of reasons why a polity might
dynamically incorporate the law of another jurisdiction (whether up,
down, or horizontally) is not meant to be exhaustive. Other grounds
can be adduced in particular circumstances. For example, states that
require their high courts to interpret their respective constitutions in
“lockstep” with the U.S. Supreme Court’s interpretation of parallel
provisions of the Federal Constitution may do so for substantive ideological reasons: because states are already bound (under the incorporation doctrine) to apply the Supreme Court’s interpretation of most
of the provisions of the Bill of Rights, a lockstep requirement can be a
means of preventing a liberal state high court from interpreting the
143
state constitution to provide for additional rights.
Or the ideological valence could be reversed. In circumstances in which state judicial
“overprotection” of rights leads to conservative results—for example,
by interpreting a state equal-protection requirement to ban affirmative action programs, regulatory takings, or gun-control laws that the
Federal Constitution would permit—a lockstep requirement will prevent a conservative state high court from invalidating liberal state
policies that the federal courts would uphold.
A lockstep requirement may not typically be understood as a form
of dynamic incorporation because the operative constitutional norm
already operates of its own force. This way of viewing the matter,
however, is misleading. The rights-protective element of a Supreme
143
For example, the Florida Legislature introduced the “lockstep” approach to
search-and-seizure cases after concern that the Florida Supreme Court was overly liberal and defense oriented. See Christopher Slobogin, State Adoption of Federal Law: Exploring the Limits of Florida’s “Forced Linkage” Amendment, 39 U. FLA. L. REV. 653, 666-73
(1987) (describing the political background to the adoption of Article 1, Section 12 of
the Florida Constitution).
2008]
Dynamic Incorporation
151
Court interpretation of the Fourteenth Amendment and the Bill of
Rights operates against the states of its own force, but the rightsdenying element does not. For example, suppose a state court would,
if left to its own devices, interpret state constitutional limits on
searches and seizures to require probable cause and a warrant before
the police may search a person’s garbage. But instead, on the authority of U.S. Supreme Court precedent interpreting the Fourth
144
Amendment, the state court holds that such police activity does not
constitute a “search.” In this situation, the state court has—in virtue
of the lockstep requirement—limited the freedom of the state’s citizens via dynamic incorporation.
This last example shows how dynamic incorporation can be used
to limit “negative” freedom, or “freedom from” government action.
Dynamic incorporation can, of course, also enhance such negative
freedom. For example, a polity that chooses to dynamically incorporate the human-rights law of a body that, over time, tends to afford
more and more protection for more and more liberties, increases its
citizens’ liberties (assuming that, absent dynamic incorporation, the
polity would not have taken such steps directly). Thus, there is no
necessary connection between dynamic incorporation and negative
liberty.
By contrast, with one important exception, dynamic incorporation
systematically reduces the “positive liberty” of a polity’s citizens by tak145
ing key decisions away from the organs of that polity. That is simply
a restatement of the point noted above in Part I: even when cost justified, dynamic incorporation undermines democratic self-government
within the polity doing the incorporating. The one exception is that
where dynamic incorporation places limits on countermajoritarian institutions within a polity—as in the example of lockstep requirements
for state constitutional provisions that might otherwise be used to invalidate internal legislation—those limits may have the effect of empowering democratic institutions within the dynamically incorporat144
See California v. Greenwood, 486 U.S. 35, 43 (1988) (holding that police inspection of garbage bags left on a curb did not constitute a Fourth Amendment
“search” requiring a warrant and probable cause, while conceding that “[i]ndividual
States may surely construe their own constitutions as imposing more stringent constraints on police conduct”).
145
On the difference between negative and positive liberty, see ISAIAH BERLIN,
Two Concepts of Liberty, in THE PROPER STUDY OF MANKIND 191, 203-06 (Henry Hardy &
Roger Hausheer eds., 1997). For a related typology, see Benjamin Constant, The Liberty of the Ancients Compared with That of the Moderns (1819), in POLITICAL WRITINGS 307 (Biancamaria Fontana ed. & trans., 1988).
University of Pennsylvania Law Review
152
[Vol. 157: 103
ing polity. As the next Part explains, there are also other potentially
important differences between dynamic incorporation of decisions by
politically accountable executive and legislative actors, on the one
hand, and dynamic incorporation of decisions by politically unaccountable judicial actors, on the other.
IV. THE POLITICAL SAFEGUARDS OF DYNAMIC INCORPORATION
Thus far, this Article has argued that dynamic incorporation of
foreign law raises a prima facie threat to democratic values, that, ceteris paribus, the threat is roughly proportional to the difficulty of revoking the decision to dynamically incorporate foreign law and that
the attendant costs and benefits of dynamically incorporating upward,
downward, or horizontally will vary greatly based on context, including the justification for dynamic incorporation. This Part addresses a
tension raised by the foregoing analysis. To attain some of the benefits of dynamic incorporation—especially the benefits of coordination
and reciprocity from multilateral upward incorporation—polities will
often need to make firm commitments, that is, to make a decision to
engage in dynamic incorporation practically irrevocable. In other
words, the benefits of some kinds of dynamic incorporation will flow
precisely when such incorporation poses the greatest threat to democratic values.
This Part first argues that political representation of the dynamically incorporating polity (or representation of its citizens) within the
polity whose laws are incorporated can compensate for the loss of local democratic control that irrevocable or nearly irrevocable dynamic
incorporation entails. For example, the EU provides for representation of its member states in the Presidency and the Commission, as
well as elections to its Parliament. Thus, when a member state agrees
to make EU law enforceable as its own law in domestic courts, it dynamically incorporates law that is, in an important sense, not fully foreign. Political safeguards at the EU level substitute for the exercise of
sovereignty at the member-state level.
This Part next asks whether this strategy—which I call the political
146
safeguards of dynamic incorporation —can be successfully employed
146
The term “political safeguards” is meant to evoke a well-known article by Herbert Wechsler, in which he suggested that the U.S. Constitution’s principal protections
against the erosion of state sovereignty could be found in the representation of the
states as political units in the federal government itself. See Wechsler, supra note 127,
at 559. The Supreme Court has relied on the notion of political safeguards to avoid
2008]
Dynamic Incorporation
153
when the body formulating law to be incorporated is a court. Does
dynamic incorporation of judge-made law pose a threat to democratic
values, and if so, can the threat be mitigated by representation within
a court? How one answers these questions may depend on how one
understands the nature of adjudication.
A. Representation as Remedy
Suppose that some number of sovereign polities wish to enter into
a mutually beneficial agreement to create a suprasovereign body that
will have the power to make laws binding on and in their respective
polities. Perhaps the contracting sovereign parties wish to create a
common economic market so as to discourage trade wars and to harmonize regulations. Or perhaps they wish to avoid the race-to-thebottom dynamic that can undermine the incentive of any single polity
to enact strong protections for workers’ rights or the natural environment. Whatever the driving force behind the quest for coordination, each polity faces a dilemma: to secure the advantages of the
agreement it must be irrevocable or nearly so, but as we saw in Part I,
the greater the obstacles to repeal of an act or treaty dynamically incorporating foreign law, the greater the threat to democracy.
Formalism poses a tempting potential solution to this problem.
An incorporating polity can decline, as a matter of internal law, to give
direct effect to the suprasovereign body’s decrees. This is not necessarily a toothless gesture. In the United States, for example, treaties—
searching judicial scrutiny of laws challenged as intruding into the sovereign prerogatives of states. See Garcia v. San Antonio Metro. Transit Auth., 469 U.S. 528, 550-51
(1985) (“[T]he composition of the Federal Government was designed in large part to
protect the States from overreaching by Congress.” (citing JESSE H. CHOPER, JUDICIAL
REVIEW AND THE NATIONAL POLITICAL PROCESS 175-84 (1980); D. Bruce La Pierre, The
Political Safeguards of Federalism Redux: Intergovernmental Immunity and the States as Agents
of the Nation, 60 WASH. U. L.Q. 779 (1982); Wechsler, supra note 127)). Decisions since
the 1990s, however, indicate a greater willingness on the Court’s part to find judicially
enforceable safeguards for federalism as well. See, e.g., United States v. Morrison, 529
U.S. 598, 617 (2000) (striking down a provision of the Violence Against Women Act
that allowed victims of gender-motivated violence to sue their attackers in federal court
because Congress may not “regulate noneconomic, violent criminal conduct based
solely on that conduct’s aggregate effect on interstate commerce”); Printz v. United
States, 521 U.S. 898, 935 (1997) (holding that the Brady Act, which required state lawenforcement officials to conduct background checks of gun buyers, offended state sovereignty); United States v. Lopez, 514 U.S. 549, 567 (1995) (invalidating the Federal
Gun-Free School Zones Act as beyond Congress’s authority under the Commerce
Clause). For a reformulation of Wechsler’s thesis emphasizing the role of state and
national political parties, see Larry D. Kramer, Putting the Politics Back into the Political
Safeguards of Federalism, 100 COLUM. L. REV. 215 (2000).
University of Pennsylvania Law Review
154
[Vol. 157: 103
whether they empower future lawmaking by a supranational body or
merely express nominally static norms—are effectively presumed to be
non-self-executing; that is, only the implementing legislation has internal effect (absent evidence to the contrary in the treaty text or per147
haps the travaux prepatoire). And in the Military Commissions Act of
2006, Congress even took the step of forbidding any person from rely148
ing directly on the Geneva Conventions for a claim of right, adding
further that in construing the implementing legislation, U.S. courts
could not even take guidance from foreign or international deci149
sions.
However, formally denying direct effect in Polity A to supranational legislation and decisions can readily backfire. Because (by hypothesis) the main point of the intersovereign agreement is to bind
the parties, the other potential signatories will not accept A’s accession to the agreement if that accession does not actually commit A to
incorporate the legislation or decisions of the suprasovereign body it
creates.
To be sure, there will be circumstances in which, as a formal matter, Polity A neither makes the suprasovereign body’s legislation or
judgments directly enforceable domestically nor renders its accession
to the intersovereign agreement formally irrevocable, but other factors may assure the remaining sovereign parties of A’s ongoing compliance. The status of the decisions of panels of the World Trade Organization (WTO) in U.S. law provides an interesting example. A
decision by a WTO panel or appellate body that some U.S. law or policy violates a treaty is not automatically effective domestically. Instead,
section 129 of the Uruguay Round Agreements Act confers discretion
on the U.S. Trade Representative, in consultation with the relevant
federal agencies—which are, of course, accountable through the
147
See Medellin v. Texas, 128 S. Ct. 1346, 1357 (2008) (noting that “none of these
treaty sources creates binding federal law in the absence of implementing legislation”);
Tel-Oren v. Libyan Arab Republic, 726 F.2d 774, 808 (D.C. Cir. 1984) (“Absent authorizing legislation, an individual has access to courts for enforcement of a treaty’s provisions only when the treaty is self-executing, that is, when it expressly or impliedly provides a private right of action.” (citing Edye v. Robertson (Head Money Cases), 112 U.S.
580, 598-99 (1884))).
148
Military Commissions Act of 2006, Pub. L. No. 109-366, § 5(a), 120 Stat. 2600,
2631 (codified at 28 U.S.C. § 2241 note (2006)).
149
Id. § 6(a)(2), 120 Stat. at 2632 (codified at 18 U.S.C. § 2441 note).
2008]
Dynamic Incorporation
155
President to the American people rather than to the WTO—to decide
150
whether to implement a WTO panel or appellate body decision.
However, a decision by the United States to defy the WTO has serious external repercussions. Most dramatically, it will lead to permission for other WTO members to impose punitive duties on U.S.
goods. The 2003 showdown between the United States and the
151
WTO was a rare example of the failure of such repercussions to deter U.S. defiance of its WTO obligations, and in the end, after the
European Community threatened sanctions that would heavily affect
swing states prior to a presidential election, the United States backed
152
down.
In nearly all other circumstances, the formal power of the
United States not to incorporate WTO rulings into its domestic law
counts for little. What matters are the practical obstacles to defiance
of the WTO, not the formal obstacles.
This is not to say that formal automatic incorporation and formal
irrevocability count for nothing. Under particular constitutional provisions and doctrines in any given polity, it may make a critical difference whether an Act or Treaty acceding to the suprasovereign body’s
authority is formally irrevocable and whether laws made by the suprasovereign body have direct effect within that polity. For example,
the United States cannot, as a formal matter, irrevocably join suprasovereign agreements for the simple reason that a later-enacted
statute approved by a simple majority vote in each house of Congress
150
See Uruguay Round Agreements Act § 129(b)(4), 19 U.S.C. § 3538(b)(4)
(2006) (providing that “[t]he Trade Representative may, after consulting with the administering authority and the congressional committees under paragraph (3), direct
the administering authority to implement, in whole or in part, the determination”
made by the agency that would be consistent with the WTO mandate).
151
In March of 2002 the United States imposed thirty-percent tariffs on certain
imported steel products and fifteen-percent tariffs on rebar and stainless steel. See
Proclamation No. 7529, 3 C.F.R. 15 (2003). Several countries promptly filed a complaint before the WTO and successfully argued that the steel tariffs violated WTO
nondiscrimination provisions. See Appellate Body Report, United States—Definitive Safeguard Measures on Imports of Certain Steel Products, WT/DS248/AB/R (Nov. 10, 2003),
available at http://www.wto.org/english/tratop_e/dispu_e/248_259_abr_e.doc.
152
The European Community (EC), aware that the steel tariffs benefited President Bush in battleground states such as West Virginia, Pennsylvania, and Ohio,
threatened economic sanctions that would affect other crucial states like Florida,
South Carolina, Washington, and North Carolina. Just prior to the retaliation deadline set by the EC, President Bush ordered an end to the steel tariffs. See Jide Nzelibe,
The Credibility Imperative: The Political Dynamics of Retaliation in the World Trade Organization’s Dispute Resolution Mechanism, 6 THEORETICAL INQUIRIES LAW 215, 224-25 (2005);
see also Rossella Brevetti & Christopher S. Rugaber, Bush Ends Steel Safeguard Tariffs in
Face of Threat by EU to Retaliate, 20 INT’L TRADE REP. 2021 (2003).
University of Pennsylvania Law Review
156
[Vol. 157: 103
(and signed by the President) takes precedence over an earlier153
enacted treaty.
For present purposes, however, we can gloss over these distinctions
and focus on irrevocability and effectiveness in practice. Regardless of
the form of the deal, what matters is whether the cost of revoking an
agreement to be bound by future enactments and judgments of a suprasovereign body is sufficiently high that, as a practical matter, it is
very unlikely that a polity, having entered the agreement, will back out.
And there’s the rub. A de facto irrevocable decision by Polity A to
enter into a suprasovereign agreement to be bound by the future enactments and judgments of some suprasovereign body poses roughly
the same threat to democratic values in Polity A as a formally irrevocable decision to do the same. In either case, important policy decisions will be made by actors that are not accountable to the citizens of
A, and if the citizens of A are unhappy about those decisions, there is
little they can do about them. Thus, formalism provides a false solution because formal revocability without actual revocability does not
mitigate the antidemocratic impact of dynamic incorporation of foreign law.
Representation is a better, and more common, solution. Consider
the situation of the states of the United States in 1787 or, more recently, the original member states of the European Economic Community (EEC) in 1957 or the EU member states when the Maastricht
Treaty came into effect in 1993. A state entering into one of these unions gave up some of its regulatory power by agreeing to accept the
larger body’s law in exchange for the reciprocal benefits of union.
But in partial compensation for the resulting loss of local democratic
control, the member states received political representation in the
larger political entity.
Representation in the larger entity does not perfectly substitute
for the lost regulatory autonomy, of course. The larger denominator
associated with a larger political entity dilutes the votes of citizens of
Polity A. If this effect is sufficiently large, citizens of a small polity may
demand special protections to prevent their interests from being
completely overrun by those of their more populous neighbors. Representation of each state by two members in the U.S. Senate, and the
assignment to each EU member state of one seat in the European
Commission, compromise the principle of “one person, one vote” at
153
E.g., Ex parte Webb, 225 U.S. 663, 683 (1912) (“Of course an act of Congress
may repeal a prior treaty as well as it may repeal a prior act.”).
2008]
Dynamic Incorporation
157
the higher political level in order to preserve some measure of selfgovernance at the lower political level. Whether such arrangements
provide sufficient representation at the suprastate level to offset the
loss of democratic accountability at the state level cannot be answered
in the abstract. But, where the benefits of the interstate agreement
are substantial, representation could be enough to tip the balance.
Certainly that was the judgment of the states that joined the United
States and of those that joined the EEC and later the EU.
Representation in the suprastate body need not be the only safeguard for states that give up some self-rule for the benefits of mutual
coordination, but it will often be essential for the enforcement of the
other safeguards. Here, the U.S. experience is especially instructive.
Perhaps the single most important protection for self-rule at the state
level is the allocation of powers among the federal and state governments. Certainly that was the belief of the original Constitution’s
154
most prominent defenders.
Yet from very early in the Republic, it
became apparent that there would be few, if any, judicially enforce155
able limits on the subject-matter competence of Congress. Notwithstanding periodic judicial efforts to rein in the scope of congressional
156
power, with the growth of the national economy, the judicially en157
forceable limits have come to play at best a marginal role.
Today,
154
As James Madison explained, “the State governments could have little to apprehend, because it is only within a certain sphere that the federal power can, in the
nature of things, be advantageously administered.” THE FEDERALIST NO. 46, at 295
( James Madison) (Clinton Rossiter ed., 1961); see also THE FEDERALIST NO. 45, at 29293 ( James Madison) (Clinton Rossiter ed., 1961) (“The powers delegated by the proposed Constitution to the federal government are few and defined. Those which are
to remain in the State governments are numerous and indefinite.”).
155
See Gibbons v. Ogden, 22 U.S. (9 Wheat.) 1, 196 (1824) (“[The power to regulate commerce,] like all others vested in Congress, is complete in itself, may be exercised to its utmost extent, and acknowledges no limitations, other than are prescribed
in the constitution.”); McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 421 (1819)
(“Let the end be legitimate, let it be within the scope of the constitution, and all means
which are appropriate, which are plainly adapted to that end, which are not prohibited, but consist with the letter and spirit of the constitution, are constitutional.”).
156
See United States v. Lopez, 514 U.S. 549, 567 (1995) (holding the Federal GunFree School Zones Act unconstitutional because it extended beyond Congress’s Commerce Clause power); Hammer v. Dagenhart (Child Labor Case), 247 U.S. 251, 277
(1918) (invalidating a federal law prohibiting the interstate shipment of goods produced by child labor), overruled by United States v. Darby, 312 U.S. 100, 115-17 (1941).
157
See Gonzales v. Raich, 545 U.S. 1 (2005) (upholding a federal statute that prohibits possessing, obtaining, or manufacturing marijuana as applied to a person growing marijuana, under state license, strictly within California for medical purposes).
University of Pennsylvania Law Review
158
[Vol. 157: 103
only one Justice of the Supreme Court argues for any serious restric158
tion.
Nonetheless, Congress is not in fact omnipotent because the “political safeguards of federalism”—that is, the role of the states and
their citizens voting on a state-by-state basis in the selection of federal
officeholders—give Congress a reason not to push its power to the full
extent that the courts would permit. Viewing the United States as a
suprastate body whose law (by virtue of the Supremacy Clause) is ir159
revocably and dynamically incorporated into the laws of the several
states, we can regard the various mechanisms by which the states are
160
represented in the federal government as mechanisms for keeping
the latter within its appropriate bounds, and thus as counterweights to
the democratic loss suffered by the states as a consequence of the dynamic incorporation. More broadly, representation can serve this
function whenever a state dynamically incorporates the law of a suprastate body.
B. Representative Courts?
In principle, a polity can dynamically incorporate future legal
rules and standards regardless of whether those rules and standards
are enacted by a legislature, promulgated by an executive agency, or
expressed in the judgment of a court. This section focuses on the
special concerns raised by using “political safeguards” to limit democracy losses when the incorporated rules or standards come from
courts.
Dynamic incorporation of judicial decisions will sometimes be
deemed a necessary element of a regime of multilateral upward incorporation. The state parties to a multilateral agreement will often
deem it useful to submit disputes concerning the meaning of provisions of the mutually incorporated law to an adjudicatory body cre-
158
See id. at 58 (Thomas, J., dissenting) (arguing for a restrictive view whereby the
Commerce Clause only “empowers Congress to regulate the buying and selling of
goods and services trafficked across state lines”); Lopez, 514 U.S. at 584-602 (Thomas,
J., concurring) (arguing that current Commerce Clause jurisprudence is “far removed
from both the Constitution and from our early case law,” and advocating a more restrictive interpretation).
159
See Texas v. White, 74 U.S. (7 Wall.) 700, 725 (1868) (describing the Union as
“perpetual” and “indissoluble”).
160
See Wechsler, supra note 127, at 543-44 (invoking the role of states in composing and selecting federal government offices as one of the chief political safeguards of
federalism).
2008]
Dynamic Incorporation
159
ated by that very agreement. Whether such an adjudicatory body acts
as a court of first instance or an appellate court, and whether it acts
directly on private parties or requires state parties to espouse the
claims of its citizens, the reason for a separate, supranational body will
usually be the same: the contracting state parties do not trust the domestic courts of the respective signatories to provide impartial justice.
Nonetheless, the mere existence of a supranational adjudicatory
body does not, as a formal logical matter, require dynamic incorporation of supranational (for our purposes, “foreign”) law. If the adjudicatory body resolves cases but does not set precedents, then its judgments are not, in a formal sense, “law” to be incorporated. Although
Anglo-American lawyers and scholars tend to regard the precedentsetting feature of adjudication as an essential guarantee of procedural
161
regularity, civil law jurisdictions have long functioned without a
formal notion of precedent. Accordingly, a regime of upwardincorporated statutory and administrative law could be devised in
which the supranational adjudicatory body did not make law that the
member states would have to incorporate.
Yet many close observers have long taken the view that, as a functional rather than a formal matter, civil law jurisdictions do have a doc162
trine of precedent.
The proliferation in recent years of transEuropean and other international courts that publish their judgments
accompanied by lengthy opinions in the “American” style—and that
adjudicate cases for both common law and civil law countries—only
further erodes the notion that high-profile adjudication can proceed
without a functional notion of precedent.
Moreover, even if it is possible to establish a system of adjudication
without precedent, it will often make policy sense to require that
161
See Payne v. Tennessee, 501 U.S. 808, 827 (1991) (“Stare decisis is the preferred
course because it promotes the evenhanded, predictable, and consistent development
of legal principles, fosters reliance on judicial decisions, and contributes to the actual
and perceived integrity of the judicial process.”); ROBERT C. POST, CONSTITUTIONAL
DOMAINS: DEMOCRACY, COMMUNITY, MANAGEMENT 30 (1995) (“If the Constitution
predominates because it is law, its interpretation must be constrained by the values of
the rule of law, which means that courts must construe it through a process of reasoning that is replicable, that remains fairly stable, and that is consistently applied.” (footnote omitted)).
162
MIRJAN R. DAMASKA, THE FACES OF JUSTICE AND STATE AUTHORITY 36-38 (1986)
(“While formally free to disregard legal opinions of their superiors, judges continued
to look to high courts for guidance.”); PETER GOODRICH, READING THE LAW 39 (1986)
(“The status of jurisprudence as law is informally recognized in that reference to previous decisions containing interpretations of the law is made during the course of legal
argument, and may be found in notes and commentaries made about the code.”).
University of Pennsylvania Law Review
160
[Vol. 157: 103
judgments of a suprastate court be given precedential effect. Suppose
one thinks that, ceteris paribus, the best way to settle disputes arising
under a supranational legal regime is for a supranational adjudicatory
body to resolve the question retrospectively for past cases, but for an
interstate legislative or administrative body to adopt new rules and
standards specifically addressing and resolving the disputed question
prospectively. Even then, the adjudicatory body will be “making law”
for past cases, and the same sorts of dysfunctions that create “gridlock” and “ossification” in domestic legislative and regulatory proc163
esses may operate at the supranational level. Thus, the adjudicatory
body’s decision may end up establishing law prospectively as well for
some considerable period of time.
Accordingly, we have reason to take seriously the possibility of a
regime of irrevocable or nearly irrevocable upward dynamic incorporation of supranational judicial lawmaking. Is such a regime a threat
to democratic values? That depends on how one understands the judicial process.
If one is committed to a strong conception of formalism, then judicial lawmaking is pure usurpation. Courts never legitimately make
law; all they do is apply preexisting legal norms. Because there will
sometimes be disagreements about how to apply those preexisting
norms and because uniform interpretation has value, there may be
sound reasons to prefer a system of supranational adjudication in
which final interpretive authority resides in a single supranational
court. There is, however, no reason in principle why the supranational court will do a better or worse job of reaching correct interpretations than would individual national courts. So long as the judges of
the supranational court are highly qualified legal technicians acting in
good faith, there will be no judicial lawmaking and thus no threat to
democratic values within the states that comprise the supranational
regime. Under this type of hyperformalism, the notion that individual
judges “represent” the states from which they hail would be a deeply
problematic solution to the problem of dynamic incorporation. But
happily for the hyperformalist, there is no problem in the first place
because there is no law to be made.
163
See SARAH A. BINDER, STALEMATE: CAUSES AND CONSEQUENCES OF LEGISLATIVE
GRIDLOCK 4-10 (2003) (finding bicameral differences, partisan polarization, and a disappearing political center to be significant causes of gridlock); Thomas O. McGarity,
Some Thoughts on “Deossifying” the Rulemaking Process, 41 DUKE L.J. 1385 (1992) (discussing causes of, and potential solutions to, the problem of regulatory “ossification”).
2008]
Dynamic Incorporation
161
For different reasons, neither is there a serious problem for the
imaginary hyperformalist’s equally imaginary antithesis, the hyperrealist. If one accepts the extreme legal-realist view that courts decide
cases no differently from political actors, then there is no reason in
principle to deny that representation can serve the same function with
respect to dynamic incorporation of judicial lawmaking as it can serve
with respect to dynamic incorporation of legislative and administrative
lawmaking—namely, it can mitigate the antidemocratic impact.
But what if one is neither a thoroughgoing formalist nor a thoroughgoing realist? What if, in other words, one thinks that law, while
not completely independent of politics, is not simply reducible to
politics either? In that case, one is likely to think that courts do legitimately make law in resolving hard cases, but also that there is
something problematic about the idea that judges “represent” constituencies in deciding among potential rules and standards.
To see whether we can fashion an acceptable role for representation in supranational judicial lawmaking, it will be helpful to consider
how representation functions in related adjudicatory contexts. Arbitration provides a stark example. In both private and public arbitration, it is common for the interested parties each to name one or
more arbitrators, with one or more “neutral” arbitrators to be selected
by a different means—commonly by the consent of the party-chosen
164
arbitrators.
Although arbitrators, once chosen, are supposed to de165
cide cases fairly, the very selection mechanism underscores the
widespread understanding that party-chosen arbitrators represent, or
at least are likely to be sympathetic to, the parties that have chosen
them. This understanding is reinforced by, and in turn reinforces,
two key procedural dimensions to most arbitration: the law as such is
less binding on decisions of arbitrators than it is on decisions of
courts, and decisions of arbitrators do not have the same precedential
166
force as decisions of courts.
164
See Stef Shipping Corp. v. Norris Grain Co., 209 F. Supp. 249, 253 (S.D.N.Y.
1962) (describing a typical tripartite arbitration agreement “where each party to a dispute is given the right to select an arbitrator and the third member is selected by them
or by a disinterested party”).
165
AAA/ABA Code of Ethics for Arbitrators in Commercial Disputes Canon I.D.
(2004) (“Arbitrators should conduct themselves in a way that is fair to all parties and
should not be swayed by outside pressure, public clamor, and fear of criticism or selfinterest.”).
166
See Shearson/Am. Express, Inc. v. McMahon, 482 U.S. 220, 259 (1987) (noting
that arbitrators have no obligation to follow precedent); Geraldine Szott Moohr, Arbitration and the Goals of Employment Discrimination Law, 56 WASH. & LEE L. REV. 395, 402-
162
University of Pennsylvania Law Review
[Vol. 157: 103
Suppose, however, that for the sorts of reasons just described, the
state parties to a multistate agreement do wish to submit disputes arising under the agreement (or laws made pursuant thereto) to adjudication that is more “law-like,” that is, to adjudication in which the judges
regard themselves as impartial and in which decisions in concrete
cases establish rules and standards of decision for other cases. Might
there still be a role for “representation” as a means of pulling the antidemocratic sting from dynamic incorporation of judge-made law?
Before considering that question, we might wonder whether there
really is any such sting when it comes to judicial lawmaking. After all,
even within a single domestic legal system, courts do not decide questions of law “democratically.” This is especially true when courts exercise the countermajoritarian power to invalidate legislative decisions
on constitutional grounds. Even when exercising the ordinary power
of statutory interpretation, courts do not typically act in an especially
democratic fashion. If, for example, the test for determining what
some act of Congress means is one that looks to the plain meaning of
the words, the context of enactment, and the legislative history, we
ordinarily think that unelected judges can do as good a job of applying
this test as elected judges. Indeed, concerns about the possibility of
corruption (due to the expense of judicial elections and the concomitant imperative of raising funds from potential parties) incline us to
think that unelected and thus disinterested judges will do a more faithful
job of applying that test than judges who represent any constituency.
The difficulty with this line of argument, however, is that it rests
on a highly formalist premise that, by hypothesis, anyone who thinks
there is a real problem here will have already rejected. To recognize
167
what we might call “the partial autonomy of law” from politics is not
to concede the formalist claim that the background, training, and values of judges make no difference at all in the resolution of concrete
cases or in judicial lawmaking. The politicians and interest groups
who treat each Supreme Court nomination as a high-stakes political
event rightly understand that such matters contribute to a “judicial
168
philosophy,” and that judicial philosophy matters.
03 (1999) (describing the nonbinding, generally confidential, and private nature of
arbitration decisions).
167
MICHAEL C. DORF, NO LITMUS TEST: LAW VERSUS POLITICS IN THE TWENTYFIRST CENTURY, at xix (2006).
168
For an excellent account of what constitutes a judicial philosophy and how it
figures in Supreme Court decision making, see CHRISTOPHER L. EISGRUBER, THE NEXT
JUSTICE: REPAIRING THE SUPREME COURT APPOINTMENTS PROCESS ch. 6 (2007). Eis-
2008]
Dynamic Incorporation
163
To be sure, someone who knows and agrees with a judge’s judicial
philosophy is not exactly “represented” by that judge. Perhaps, however, supranational judicial lawmaking does not require anything
quite so strong as representation in order to serve the same function
as representation does in the context of supranational legislative and
administrative lawmaking. That function might be better described as
something like “values alignment.”
Domestically, when Presidents, senators, and the interested public
seek judges and Justices who share their values, they do not (if acting
in good faith) seek judges and Justices who will literally take orders.
Indeed, even an elected representative—according to Edmund Burke,
for example—does not simply take instruction from his constitu169
ents.
But the distance is greater for a judge than for a legislator.
We think it laudable when a legislator listens to the concerns of her
constituents or, in running for office, explains what policies she proposes to pursue. Supreme Court nominees, by contrast, pointedly refuse to answer questions that even hint at how they will vote, if confirmed. Much of this, of course, is a self-serving pose—what Senator
170
Joseph Biden aptly called a “Kabuki dance” —because a prospective
Justice could say a great deal more than recent nominees have said,
while still promising to keep an open mind to new arguments if confirmed. The fact that Justices do not take this approach has more to
do with nomination and confirmation politics than with judicial eth171
ics.
Nonetheless, it is hard to imagine anyone seriously maintaining
that judicial nominees should make campaign promises in the same
gruber describes a judicial philosophy as the sum of a judge’s “ideological convictions”
and “procedural convictions, including . . . convictions about the proper role of courts
within the American political system.” Id. at 99.
169
Burke believed that a representative should exercise his best judgment about
the public interest, even if that means disappointing his constituents. See Edmund
Burke, Speech to the Electors of Bristol on Being Elected (Nov. 1774), in THE POLITICAL PHILOSOPHY OF EDMUND BURKE 108, 110 (Iain Hampsher-Monk ed., 1987). He
successfully ran for Parliament and put these principles into practice, whereupon he
was voted out of office. ROBERT LUCE, LEGISLATIVE PRINCIPLES: THE HISTORY AND
THEORY OF LAWMAKING BY REPRESENTATIVE GOVERNMENT 464 (1930).
170
Confirmation Hearing on the Nomination of John G. Roberts, Jr. to Be Chief Justice of the
United States: Hearing Before the S. Comm. on the Judiciary, 109th Cong. 328 (2005) [hereinafter Confirmation Hearing] (statement of Sen. Joseph R. Biden, Jr., Member, S.
Comm. on the Judiciary).
171
Christopher Eisgruber argues that confirmation hearings are most likely to be
acrimonious, and thus that nominees are most likely to try to hide their views, when
Presidents nominate ideological extremists rather than moderates. See EISGRUBER, supra note 168, ch. 7.
University of Pennsylvania Law Review
164
[Vol. 157: 103
way that candidates for legislative or executive office do. If judges
“represent” particular viewpoints within a domestic legal system (or at
least within the one domestic legal system with which I am most familiar) only in the limited sense that citizens have a right to demand that
their senators only confirm judges who broadly share their values and
hold a judicial philosophy they deem acceptable, then the appropriate
question for supranational lawmaking courts is whether they can be
made representative in roughly the same way.
The answer to that question is probably yes. Today, Presidents
rarely pay much attention to the state or region from which a prospective Supreme Court nominee hails, except perhaps as a means of placating a particular senator or as a proxy for particular values (to the
extent that “red-state, blue-state” divisions translate into judicial philosophy). By contrast, ideology and (sometimes) membership in a
particular racial or ethnic group play substantial roles in presidential
nomination and Senate confirmation. In previous eras, however,
when political cleavages closely tracked geographic lines, political actors paid close attention to issues of state and regional balance on the
172
Court. Although no one thought that a southern Justice was legally
or even morally bound to represent southern interests on questions
involving slavery in quite the way that southern state delegations to
Congress did, it was understood that geographical origin shaped outlook.
Likewise, if we look at prominent international judicial bodies today, we find that membership criteria pay exquisite attention to the
geographic area from which judges are selected. For example, the International Court of Justice (ICJ) at The Hague, the principal judicial
organ of the United Nations, “may not include more than one national of the same State. Moreover, the Court as a whole must represent the main forms of civilization and the principal legal systems of
173
the world.” The formula is even more rigid for the European Court
of Human Rights (ECHR) in Strasbourg. Although “[j]udges sit on
the Court in their individual capacity and do not represent any
172
See HENRY J. ABRAHAM, JUSTICES, PRESIDENTS, AND SENATORS: A HISTORY OF
U.S. SUPREME COURT APPOINTMENTS FROM WASHINGTON TO CLINTON 3 (new &
rev. ed. 1999) (including “geographic” representation in a list of various representative
factors considered in judicial appointments).
173
Int’l Court of Justice, Members of the Court, http://www.icj-cij.org/court/
index.php?p1=1&p2=2 (last visited Oct. 1, 2008).
THE
Dynamic Incorporation
2008]
165
174
State,” the total number of judges equals the number of members of
175
the Council of Europe, with each Council member having the right
176
to put forward names for one seat on the court.
The greater diversity of viewpoints likely to be found in a global
organization relative to a national one (or, in a federal system, the
greater viewpoint diversity to be found at the national level than at the
state level) makes it more difficult for an international (or national)
court to achieve rough viewpoint representativeness than for a national (or state) court to achieve such representativeness. Simply as a
practical matter, for a court to function collegially requires some relatively small number of judges—nine, eleven, or perhaps fifteen, but
not substantially larger—or what amounts to the same thing: a practice of hearing cases in panels rather than in plenum.
This practical limit in turn means that many member states of a
large supranational judicial system will have to settle for virtual representation. For example, Nicaragua and Colombia count on judges
from Venezuela and Mexico to represent, if not necessarily their interests, at least their way of understanding their legal obligations. Indeed, given that the ICJ’s jurisdiction includes the resolution of terri177
torial disputes, which are most likely to arise between nations in the
same region, it should be clear that regional representation is keyed
not to representation of interests but to understanding of perspective.
But if member states of a multistate court must settle for virtual
representation, it is not clear that constituencies do much better under the selection systems in place for national courts. The United
States has perhaps the world’s most highly politicized process for
choosing the judges of its constitutional court, and one would think
that the system would therefore routinely produce moderate Justices
broadly in tune with national values. In fact, however, depending on
the happenstance of Senate control and a President’s commitment to
change through the courts, a small difference in elections can result
174
COUNCIL OF EUROPE, EUROPEAN COURT OF HUMAN RIGHTS, SURVEY OF ACTIVI2007, at 5 (2008), available at http://www.coe.int/t/dc/files/themes/cedh/
1-2248184-Survey_of_activities.pdf.
175
See id. (“The Court is composed of a number of judges equal to that of the
Contracting States (currently forty-five).” (footnote omitted)). There are actually
forty-seven members, but Ireland and Montenegro’s seats are currently vacant. See id.
at 5 n.6.
176
For a list of members as of December 31, 2007, see id. at 8.
177
See, e.g., Territorial and Maritime Dispute (Nicar. v. Colom.), 2007 I.C.J. 124
(Dec. 13), available at http://www.icj-cij.org/docket/files/124/14305.pdf (accepting
jurisdiction over a territorial dispute).
TIES:
University of Pennsylvania Law Review
166
[Vol. 157: 103
in substantial philosophical differences among the Justices that are
confirmed under different administrations. In a supranational system
in which judicial seats are apportioned roughly one to a country or
region, one can at least count on one judge with a broadly sympathetic perspective, and norms of neutrality and collegiality may then
give that judge substantial influence. Thus, the virtual representation
for which many nations must settle in the selection of judges for supranational courts may do a better job of reflecting the interests and
outlooks of the member nations than a more political, and thus more
directly representative, system would do.
To be clear, the principal example of “representation” on supranational judicial bodies I have used here—the ICJ—does not issue
judgments that are automatically or irrevocably incorporated into the
domestic law of UN member states. Thus, for example, the U.S. Supreme Court held in Sanchez-Llamas v. Oregon that ICJ rulings interpreting the Vienna Convention on Consular Relations were entitled to
“respectful consideration,” but were not binding on the Supreme
Court, which went on to disagree (respectfully) with the ICJ’s conclu178
sions.
Meanwhile, even before the Court’s decision in SanchezLlamas, the State Department had given notice that it would prospectively withdraw from the Vienna Convention’s Optional Protocol,
which had invoked the ICJ’s compulsory jurisdiction in the first
179
place.
If representation of member states among the judges of a supranational judicial body is thought to be a necessary expedient when the
supranational judicial body’s judgments lack the automatic force of
law within the member states, such representation would seem to be
especially necessary where the supranational court’s judgments have
greater force. Whether such representation is a sufficient condition to
compensate for the loss of domestic judicial authority is a different
question, but the answer is probably yes. Representation on courts accomplishes less than representation in legislative and administrative
bodies. At the same time, however, delegating judicial decisions to a
supranational court poses less of a threat to domestic democracy because judicial decision making itself is not best understood as a democratic process. Thus, to compensate for irrevocable or nearly irrevocable dynamic incorporation of judge-made law, judicial
178
179
548 U.S. 331, 355-58 (2006).
See id. at 339.
2008]
Dynamic Incorporation
167
representation can succeed despite accomplishing less than legislative
or administrative representation.
Before concluding this Part, it is worth noting an argument that is
not available to justify nearly irrevocable dynamic incorporation of
judge-made law. In a recent article, Henry Monaghan argues that supranational judicial review does not violate U.S. constitutional
180
norms.
Monaghan’s best arguments are historical. The United
States, he contends, has submitted to supranational adjudication since
181
the early days of the Republic.
Monaghan also advances a functional argument, however. Supranational adjudication, he argues, is
small potatoes compared to supranational legislation. As to the latter,
he concedes that “[i]ssues of national sovereignty and democratic accountability are surely raised by this increasingly widespread practice.
182
But that bell having been rung, it cannot be unrung.”
In other
words, because the United States has acquiesced in the sovereigntythreatening practice of supranational legislation, it is too late to worry
about the lesser threat posed to national sovereignty by supranational
adjudication.
However, this line of argument is unavailable here because, as
Monaghan’s examples themselves reveal, he conceptualizes supranational adjudication as a form of arbitration or other dispute-resolution
mechanism that does not create law. Indeed, he is careful to distinguish these mechanisms not from legislation as such, but from the
183
broader category of “lawmaking.”
Presumably, supranational judicial lawmaking poses the same serious threat to national sovereignty as
184
other forms of supranational lawmaking.
To be sure, Monaghan also thinks that even these more serious
threats have been accepted as the price of participation in the global
180
See generally Monaghan, supra note 3.
See id. at 851-66 (detailing how the United States has entered international
agreements creating state-state arbitration panels to resolve the private law claims of its
nationals against foreign governments, sometimes leading to a reexamination of decisions of the Supreme Court itself).
182
Id. at 882.
183
See id. at 881-82.
184
Like almost everything else, the question of whether a supranational tribunal
makes law will not always be amenable to a yes-or-no answer. The decisions of tribunals that, in theory, only resolve disputes between parties may have varying degrees of
precedential effect in practice. Cf. Robert B. Ahdieh, Between Dialogue and Decree: International Review of National Courts, 79 N.Y.U. L. REV. 2029, 2034 (2004) (describing the
NAFTA Chapter 11 review process and offering “dialectical review” as an intermediate
option between nonbinding “dialogue” among national and supranational tribunals
and strictly binding judicial review).
181
University of Pennsylvania Law Review
168
[Vol. 157: 103
order. I am less certain on this point, perhaps because I am less persuaded than he that one of the safeguards that he regards as crucial
still exists—“the ultimate prerogative of the political branches to
185
withdraw from any supranational institution.”
As a formal matter,
the Supremacy Clause of the U.S. Constitution does ensure that, absent constitutional amendment, the political branches do retain that
ultimate prerogative. Yet, as I argued in the previous section, those
who are concerned about sovereignty should be concerned principally
about whether the delegation of lawmaking power is functionally
186
revocable, and not merely about whether it is formally revocable.
Thus, supranational judicial lawmaking and other forms of supranational lawmaking pursuant to functionally irrevocable grants of
power do pose threats to sovereignty. Whether those threats rise to
the level of constitutional violations in the United States or any other
country is not my main concern here. Rather, as I have endeavored to
show throughout this Part, the functional irrevocability of a delegation
to a supranational lawmaking body—whether that body is nominally
legislative, administrative, or judicial—calls for safeguards to mitigate
the effect on democratic values. Where the delegation is otherwise
sensible, representation in the supranational lawmaking bodies themselves is the most natural such safeguard, and though one needs to
stretch the notion of “representation” somewhat to accommodate judicial lawmaking, given the other differences between judicial and
nonjudicial lawmaking, the solution works reasonably well for judicial
lawmaking also.
CONCLUSION
In recent years, much judicial, legislative, and academic attention
has been paid to the question of whether it is appropriate for U.S.
187
courts to cite foreign law as guidance in interpreting U.S. law.
185
Monaghan, supra note 3, at 882.
Supra Part IV.A.
187
See, e.g., Military Commissions Act of 2006, 18 U.S.C. § 2441 note (“No foreign
or international source of law shall supply a basis for a rule of decision in the courts of
the United States in interpreting the prohibitions enumerated in [18 U.S.C. § 2441(d),
the War Crimes Act].”); Confirmation Hearing, supra note 170, at 25 (statement of Sen.
Mike DeWine, Member, S. Comm. on the Judiciary) (“Many are troubled when they
see the Court cite international law in its decisions . . . .”). Compare Roper v. Simmons,
543 U.S. 551, 578 (2005) (“It is proper that we acknowledge the overwhelming weight
of international opinion against the juvenile death penalty . . . .”), and id. at 604-05
(O’Connor, J., dissenting) (acknowledging some role for international law in Eighth
Amendment jurisprudence because “this Nation’s evolving understanding of human
186
2008]
Dynamic Incorporation
169
Much of the criticism of this practice misses the mark, for those who
favor it never contend that foreign judgments should be treated as
188
binding precedents. But if misapplied as a criticism of citation practices, the points made in opposition to the invocation of foreign law
have greater force in circumstances in which the law of one jurisdiction dynamically incorporates the law of another jurisdiction. Regardless of whether that other jurisdiction contains, is contained within, or
is wholly outside of the incorporating jurisdiction, the act of incorporation cedes some measure of political accountability to actors answerable to a different polity. This practice can sometimes be justified
by the benefits it confers, and the concerns it raises can be mitigated
by a variety of procedural mechanisms, but the concerns cannot simply be dismissed.
dignity certainly is neither wholly isolated from, nor inherently at odds with, the values
prevailing in other countries”), with id. at 622-28 (Scalia, J., dissenting) (“[T]he basic
premise of the Court’s argument—that American law should conform to the laws of
the rest of the world—ought to be rejected out of hand.”). For an argument for “the
emergence of a transnational law . . . that merges the national and the international,”
see Harold Hongju Koh, International Law as Part of Our Law, 98 AM. J. INT’L L. 43, 53
(2004). For an argument that “[f]oreign and international law cannot be legitimately
used in an outcome-determinative way to decide questions of constitutional interpretation,” see Robert J. Delahunty & John Yoo, Against Foreign Law, 29 HARV. J.L. & PUB.
POL’Y 291, 296 (2005).
188
Justice Kennedy stated for the Court in Roper v. Simmons that foreign experience merely confirmed the conclusions that he and his colleagues would have otherwise reached independently: “The opinion of the world community, while not controlling our outcome, does provide respected and significant confirmation for our own
conclusions.” 543 U.S. at 578.